International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org

Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856

A Two Warehouse Inventory Model for

Deteriorating Items with Shortages under
Genetic Algorithm and PSO
1

Dr. Ajay Singh Yadav,2Dr. Kusum Gupta, 3Mr. Ankur Garg

, 4Ms. Anupam Swami

SRM University NCR Campus, GZB.

Banasthali University Jaipur, Rajasthan

3

MIET College Meerut U.P.

Govt. P. G. College Sambhal, U.P.

Abstract
The purpose of this article is to evaluate the value of
integrating warehouse and inventory decision. Therefore in
this paper a two warehouse inventory model for deteriorating
items is considered under assumption that the Inventory cost
(including holding cost and deterioration cost) in RW (Rented
Warehouse) is higher than those in OW (Owned Warehouse)
due to better preservation facilities in RW. The demand and
holding cost, both are taken as a variable. Shortages are
allowed in the OW fully backlogged at the next replenishment
cycle .This paper mainly dealt with deteriorating items with
linear demand and variable holding cost which is constant for
a definite time period and after that it increases linearly
according to a function of ordering cycle length in RW and
remains constant in OW. Transportation cost is incurred and
is function of distance irrespective of number of items and the
capacity of vehicle. Four kinds of meta-heuristic algorithms
or meta-heuristic hybrid algorithms: harmony search, particle
swarm optimization, genetic algorithms, simulated annealing.
Computational results showed that HS, hybrid of HS and
hybrid method of HS performed the best, based on objective
function values respect to other kind of the algorithms. Other
applications of HS in inventory and supply chain
managements can be extended to consider pricing problems. A
numerical example is presented to illustrate and validate the
model and sensitivity analysis is carried out in case-1for
different values of a parameter keeping rest unchanged.

Keywords:- Two warehouses, shortages,

algorithms, particle swarm optimization,
algorithms,

hybrid
genetic

I. Introduction
In the past, researchers have established a lot of research
in the field of Inventory management and Inventory
control system. Inventory management and control system
basically deals with demand and supply chain problems.
In the traditional models it is assumed that the demand
and holding cost are constant and goods are supplied
instantly under infinite replenishment policy when
demanded but as time passed away many researchers
considered that demand may vary with time, due to price,
time and on the basis of other factors and holding cost
also may vary depending on some factors. Many models
have been developed considering various time dependent

Volume 4, Issue 5(2), September October 2015

demand with shortages and without shortage. All those

models that consider demand variation in response to
inventory level, assume that the holding cost is constant
for the entire inventory cycle. In studies of inventory
models, unlimited warehouse capacity is often assumed.
However, in busy marketplaces, such as super markets,
corporation markets etc. the storage area for items may be
limited. Another case, of inadequate storage area, can
occur when a procurement of a large amount of items is
decided. That could be due to, an attractive price discount
for bulk purchase which is available or, when the cost of
procuring goods is higher than the other inventory related
costs or, when demand for items is very high or, when the
item under consideration is seasonal product such as the
yield of a harvest or, when there are some problems in
frequent procurement. In this case these items cannot be
accommodated in the existing store house (the own
warehouse, abbreviated as OW). Hence, in order to store
the excess items, an additional warehouse (the rented
warehouse, abbreviated as RW),which may be located at a
short distance from the OW or a little away from it, due to
non-availability of ware-house nearby, is hired on a rental
basis.

II. Genetic Algorithms

The basic concepts were developed by Holland, while the
practicality of using the GA to solve complex problems
was demonstrated. Genetic Algorithms (GAs) is a soft
computing approach.
GA is a maximization process. The problem it addresses
usually has a very large search space with probable
multiple local maxima inside it. The GA process has to
ensure that it is not trapped at local maxima, so that at the
end of the process it may find the global maxima even if
the global maximum is not returned, we may expect a
close approximation of it as the outcome of the GA
process
To achieve this, GA works on a set of solutions to the
given problem instance, and evolves it through a number
of generations. The evolution process stops when some
predefined termination condition is satisfied. At each
intermediate stage, the old generation is replaced by the
Page 40

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
new generation. The individuals of the population of a
generation are processed with the help of a number of GA
operators in such away that the quality of the new
generation, in general is improved in comparison with old
generation. In this way we obtain better and better
solution will be returned by the GA process.
Procedure Basic GA
Step 1:- Initialize the population. Call this the current
population.
Step 2:- Repeat step 3 through step 5 till termination
condition is satisfied.
Step 3:- Apply selection operation on the current
population to obtain the mating pool.
Step 4:- Apply crossover and mutation operators n for the
matting pool to generate the new Population.
Step 5:- Replace the current population by the new
population.
Step 6:- Return the best solution of the current population.

experience and that of other particles present in the

swarm. The algorithm starts with a population (swarm) of
random solutions (particles). Each particle is assigned a
random velocity and allowed to move in the problem
space. The particles have memory and each of them keeps
track of its previous (local) best position.
Particles velocity
Particles position
b
number of elements in a particle,
q
Inertia weight of the particle,
m generation number,
a1 , a2
acceleration constants,
e and
random value between 0 and 1
Local best position of the particle,
Global best position of particle in the swarm

Concept of modification of a searching point by PSO

IV. Hybrid Techniques

Figure1:- GA flow chart

III. Particle Swarm Optimization (PSO)

Particle Swarm Optimization (PSO) introduced by
Kennedy and Eberhart in 1995 is a population based
evolutionary computation technique. It has been developed
by simulating bird flocking fish schooling or sociological
behaviour of a group of people artificially. Here the
population of solution is called swarm which is composed
of a number of agents known as particles. Each particle is
treated as point in d-dimensional search space which
modifies its position according to its own flying

Volume 4, Issue 5(2), September October 2015

Hybrid techniques are the integration of various soft

computing techniques that synergizes the strength and
weakness of an individual technique with the other
therefore the overall performance of such techniques
working in unison is increased.
The three main pillars of soft computing are fuzzy logic,
artificial neural networks and evolutionary search
particularly genetic algorithms -all of these have been
successfully applied in isolation to solve practical
problems in various fields where conventional techniques
have been found inadequate. However these techniques are
complementary to each other and they may work
synergistically combining the strengths of more than one
of these techniques as the situation may occasionally
demand. Hybrid systems are systems where several
techniques are combined to solve a problem. Needless to
say that such amalgamation must be used only if it returns
better results than any of these techniques in isolation.
Based on how or more systems are combined hybrid
systems have been classified into three broad categories
Page 41

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
viz sequential, auxiliary and embedded hybrid systems.
Three primary types of hybrid systems are briefly
discussed in this paper these are neuro-genetic, neurofuzzy and fuzzy-genetic hybrid systems. Hybrid systems
integrating all three techniques, viz, fuzzy logic neural
network, and genetic algorithms are also implemented in
several occasions.

V. Simulated Annealing
Simulated Annealing (SA) is a technique for finding good
solution to minimization problems. It simulates the
physical annealing process of solidifying a metal to a
uniform crystalline structure. In order to achieve this
uniform crystalline structure the metal is first heated to a
molten state and the gradually cooled down. The critical
parameter of this process is the rate of cooling. If the
cooling takes place too quickly energy gaps will be formed
resulting in non-uniformity in the crystalline structure. On
the other hand if the cooling takes place too slowly then
time is wasted. The optimal cooling rate varies from metal
to metal.

VI. Related Work

Hartely (1976) discussed an inventory model with two
storage facilities. Ghare and Schrader (1963) initially
worked in this field and they extended Harris (1915) EOQ
model with deterioration and shortages. Goyal and Giri
(2001) gave a survey on recent trends in the inventory
modelling of deteriorating items. Lee and Wu (2004)
developed a note on EOQ model for items with mixtures
of exponential distribution deterioration, shortages and
time varying demand. It is generally assumed, that the
holding cost in the RW is higher than that in the OW, due
to the additional cost of maintenance, material handling,
etc. To reduce the inventory costs, it will be cost-effective
to consume the goods of the RW at the earliest. Assuming
the deterioration in both warehouses, Sarma (1987),
extended his earlier model to the case of infinite
replenishment rate with shortages. Pakkala and Achary
(1992) extended the two-warehouse inventory model for
deteriorating items with finite replenishment rate and
shortages, taking time as discrete and continuous variable,
respectively. In these models mentioned above the demand
rate was assumed to be constant. Subsequently, the ideas
of time varying demand and stock dependent demand
considered by some authors, such as Goswami and
Chaudhary (1998), Bhunia and Maiti (1998), Bankerouf
(1997), Kar et al. (2001) and others. Bhunia and Maiti
(1994) extended the model of Goswami andChaudhary
(1972), in that model they were not consider the
deterioration and shortages were allowed and backlogged.
Goel and Giri (2001) suggested a review of deteriorating
inventory literature in which all the inventory models for
deteriorating items assume that the deterioration occurs as
soon as the retailer receives the commodities.
Last so many years, some researchers have given attention
to the situation where holding cost is taken constant per
item per unit of time but due to the time value of money it
is not assumed that holding cost will remain always
constant and thus holding cost may vary with time. Chang

Volume 4, Issue 5(2), September October 2015

(2004) developed an inventory models with stock

dependent demand and nonlinear holding costs for
deteriorating items. Ajanta Roy (2008) developed an
inventory model for deteriorating items with time varying
holding cost and price dependent demand. C. Sugapriya
and K. Jeyaraman (2008) developed an EPQ model for
deteriorating items in which holding cost varies with time.
Maya Gyan and A.K. Pal (2009) developed a two ware
house inventory model for deteriorating items with stock
dependent demand rate and holding cost. BinduVaish and
Garima Garg (2011) consider variable holding cost for
development of Optimal Ordering and Transfer Policy for
an Inventory System. Mukesh Kumar et.al. (2012)
developed A Deterministic Inventory Model for
Deteriorating Items with Price Dependent Demand and
Time Varying Holding Cost under Trade credits. Yadav
A.S. and Swami A. (2013) developed a two ware house
inventory model for deteriorating items with exponential
demand and variable holding cost. K.D. Rathor and P.H
Bhathawala [2013]constructed a model with variable
holding cost and inventory level dependent demand. R. P.
Tripathi [2013] developed an Inventory model for varying
demand and variable holding cost.Vinod Kumar Mishra
et.al. (2013) developed an inventory model with variable
holding cost and salvage value. Vipin Kumar et. el. (2013)
developed an inventory model with selling price
dependent demand and variable holding cost.
Taleizadeh AA, et.al. (2008). developed Optimizing
Multi-Product Multi-Constraint Inventory Control
Systems with Stochastic Replenishment and an EOQ Joint
Replenishment Policy to Supply Expensive Imported Raw
Materials with Payment in Advance, multi products, multi
constraints newsboy problem with batch order and
discount.
Taleizadeh AA, et.al (2009) extended A Particle Swarm
Optimization Approach for Constraint Joint Single BuyerSingle Vendor Inventory Problem with Changeable LeadTime and (r,Q) Policy in Supply Chain. Developing a
Multiproducts Newsboy Problem with a Hybrid Method of
Fuzzy Simulation and Simulated Annealing. A Hybrid
Method of Pareto, TOPSIS and Genetic Algorithm to
Optimize Multi-Product Multi-Constraint Inventory
Control Systems with Random Fuzzy Replenishments.
Optimizing multi product multi constraints Bi-objective
newsboy Problem with discount by Hybrid Method of Goal
Programming and Genetic Algorithm. A Hybrid Method
to Optimize Bi-Objective Single Period Newsboy Problem
with Fuzzy Cost and Incremental Discount. Multi-product
multi-constraint inventory control systems with stochastic
replenishment and discount under Fuzzy Purchasing Price
and Holding Costs. Multi-product multi-constraint
inventory control systems with stochastic period length
and total discount under fuzzy purchasing price and
holding costs. Optimizing multi-product multi-constraint
bi-objective newsboy problem with discount by hybrid
method of goal programming and genetic algorithm. A
harmony search algorithm to optimize chance constraint
stochastic inventory systems with dynamic demand. Joint
replenish-up-to multi products EOQ with fuzzy rough
Page 42

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
demand. N. Jawahar and A.N. Balaji (2009) developed A
genetic algorithm for the two-stage supply chain
distribution problem associated with a fixed charge.
Haixin Zhang et.al. (2013) suggested A modified multicriterion optimization genetic algorithm for order
distribution in collaborative supply chain. S.P.
Nachiappan and N. Jawahar (2007) developed a genetic
algorithm for optimal operating parameters of VMI
system in a two-echelon supply chain. Z.H. Che and C.J.
Chiang (2010) suggested A modified Pareto genetic
algorithm for multi-objective build-to-order supply chain
planning with product assembly. Keyvan Sarrafha et.al.
(2015) developed A bi-objective integrated procurement,
production, and distribution problem of a multi-echelon
supply chain network design: A new tuned MOEA. Ata
Allah Taleizadeh et.al. (2011) extended Multiple-buyer
multiple-vendor multi-product multi-constraint supply
chain problem with stochastic demand and variable leadtime: A harmony search algorithm. Wei-Chang Yeh and
Mei-Chi Chuang (2011) suggested Using multi-objective
genetic algorithm for partner selection in green supply
chain problems. S.A. Torabi, et.al. (2006) extended A
hybrid genetic algorithm for the finite horizon economic
lot and delivery scheduling in supply chains. Pavan
Kumar Naraharisetti et.al. (2007) suggested Optimal
supply chain redesign using genetic algorithm. MoonChan Kim et.al. (2008) developed Forwardbackward
analysis of RFID-enabled supply chain using fuzzy
cognitive map and genetic algorithm. Partha Guchhait
et.al. (2013) extended A production inventory model with
fuzzy production and demand using fuzzy differential
equation:
An
interval
compared genetic
algorithm approach. U.K. Bera et.al. (2012) extended
Inventory model with fuzzy lead-time and dynamic
demand over finite time horizon using a multiobjective genetic algorithm. Ata Allah Taleizadeh et. al.
(2009) suggested A hybrid method of Pareto, TOPSIS
and genetic algorithm to optimize multi-product multiconstraint inventory control systems with random fuzzy
replenishments. Ali Roozbeh Nia et.al. (2014) extended A
fuzzy vendor managed inventory of multi-item economic
order quantity model under shortage: An ant colony
optimization algorithm. Javad Sadeghi et. al. (2014)
extended
optimizing
a
hybrid
vendormanaged inventory and transportation problem with fuzzy
demand:
An
improved
particle
swarm
optimization algorithm. Manas Kumar Maiti (2011)
developed A fuzzy genetic algorithm with varying
population size to solve an inventory model with creditlink ed promotional demand in an imprecise planning
horizon. H. Altay Guvenir and Erdal Erel (1998)
suggested Multicriteria inventory classification using
a genetic algorithm. Ali Diabat (2014) extended
Hybrid algorithm for a vendor managed inventory system
in a two-echelon supply chain. K.L. Mak et. al. (1999)
developed optimal inventory control of lumpy demand
items using genetic algorithms. A.K. Maiti et. al. (2006)
extended
An
application
of
real-coded genetic
algorithm (RCGA) for mixed integer non-linear

Volume 4, Issue 5(2), September October 2015

programming in two-storage multi-item inventory model

with discount policy. Kuo-Ping Lin et.al. (2010)
developed
A
simulation
of
vendor
managed inventory dynamics using fuzzy arithmetic
operations with genetic algorithms. Sung-Pil Hong and
Yong-Hyuk Kim (2009) suggested A genetic algorithm for
joint replenishment based on the exact inventory cost.
Manas Kumar Maiti and Manoranjan Maiti (2007)
extended Two-storage inventory model with lot-size
dependent fuzzy lead-time under possibility constraints
via genetic algorithm. Arindam Roy et. al. (2009)
developed a production inventory model with stock
dependent demand incorporating learning and inflationary
effect in a random planning horizon: A fuzzy genetic
algorithm with varying population size approach. S.M.
Disney
et.
al.
(2000)
suggested
Genetic
algorithm optimisation of a class of inventory control
systems.
For last so many years, no researchers have given
attention to the situation about advertisement cost and
hence not taken into account but now days the
advertisement is most important phenomena of marketing
and business. So companies invested a part of his business
investment on advertisement for better sell in competitive
market and hence it is very important to give attention
about this phenomenon and to apply to reduce the total
inventory cost and maximize the profit.
In this paper a deterministic Inventory model for
deteriorating items with two level of storage system and
time dependent demand with fully backlogged shortages is
developed. Stock is transferred RW to OW under bulk
release pattern and the transportation cost is taken into
account and is a function of distance. The deterioration
rates in both the warehouses are constant but different due
to the different preservation procedures. Holding cost is
considered to be constant up to a definite time and is
increases according to a function of ordering cycle length.
To solve the models, four kinds of meta-heuristic
algorithms or meta-heuristic hybrid algorithms: harmony
search, particle swarm optimization, genetic algorithms,
simulated annealing. Computational results showed that
HS, hybrid of HS and hybrid method of HS performed the
best, based on objective function values respect to other
kind of the algorithms. Other applications of HS in
inventory and supply chain managements can be extended
to consider pricing problems. The numerical example is
presented to demonstrate the development and to validate
the model. Sensitivity analysis is performed separately for
each parameter.
VII. Assumption and notations
The mathematical model of two warehouse inventory
model for deteriorating items is based on the following
notation and assumptions.
Notations:
Oc : Ordering cost per Order
W : Capacity of OW.
S : Capacity of RW.
T : The length of replenishment cycle.
Mmax : Maximum Inventory level per cycle to be ordered.
Page 43

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
t1 : The time up to which inventory vanishes in RW.
t2 : The time at which inventory level reaches to zero in
OW and shortages begins.
td : Definite time up to which holding cost is constant.
hw : The holding cost per unit time in OW .
hr : The holding cost per unit time in RW
sc : The shortages cost per unit per unit time
dr : Deterioration cost in RW
dw : Deterioration cost in OW
Ac : Advertisement cost
Ir(t) : The level of inventory in RW
Ii(t) : The level of inventory in OW where i=1,2.
Is(t) : Determine the inventory level at time t in which
the product has shortages.
: Deterioration rate in RW Such that 0<<1;
: Deterioration rate in OW such that 0<<1;
d
: Distance from RW to OW
TC
: Transportation cost given by
where
Cp : Purchase cost per unit of items.
IB : Inventory backlogged.
CS : The present worth cost of shortages
CH : The present worth cost of holding inventory
(
: The total relevant inventory cost per unit
time of inventory system.
Assumption
1 Replenishment rate is infinite and lead time is
negligible i.e. zero.
2 The time horizon of the inventory system is infinite.
3 Goods of OW are consumed only after the
consumption of goods kept in RW due to the more
holding cost in RW than in OW.
4 The OW has the limited capacity of storage and RW
has unlimited capacity.
5 Demand vary with time and is linear function of time
and
given
by
D(t)=

backlogged(Mmax-IB) units are cleared and the remaining

units M is kept into two storage as W1 units in OW and
W2 units in RW.(See Figure-1)
During the time interval [0 t1] the inventory in RW
decrease due to the demand and deterioration and is
governed by the following differential equation:
In the time interval [t1 t2] the inventory level decreases in
OW decreases due to deterioration only and is governed by
differential equation
In the time interval [t1t2] the inventory level is decreases
due to demand and deterioration both and is governed by
the following differential equation

Now at t=t2 the inventory level vanishes and the shortages

occur in the time interval [t2T] and shortages is fully
backlogged and the shortages quantity supplied to the
customers at the beginning of the next replenishment
cycle. The shortages is governed by the differential
equation
At the time t=T replenishment cycle restarts. The
objective of the model is to minimize the total inventory
cost by the relevant cost as low as possible.
Now inventory level at different time intervals is given by
solving the above differential equations (1) to (4) under
boundary conditions

therefore Differential eq. (1) gives

; wherea>0 and b>0 ;

6 For deteriorating items a fraction of on hand

inventory deteriorates per unit time in both the
warehouse with different rate of deterioration.
7 Shortages are allowed and demand is fully
backlogged at the beginning of next replenishment
cycle.
8 The unit inventory cost (Holding cost) in RW>OW.
9 We assume that the holding cost will be fixed till a
definite time in RW and the will increased according
to a fraction of ordering cycle length. So for holding
cost (hr) ,we have
a time moment before which
holding cost is constant.

Now at

Ir(0)= S therefore equation (5) yield

Maximum amount of inventory backlogged during

shortages period( at t=T) is given by
The maximum Inventory to be ordered is given as

Mathematical formulation of model and analysis

In the beginning of the cycle at t=0 a lot size of Mmax units
of inventory enters into the system in which

Volume 4, Issue 5(2), September October 2015

Page 44

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
Table-1:

Figure-3 Convexity of the Model for the case-1 is

represented graphically with optimal values of
and
*.

Case-2

The optimal values of if , and T can be calculated after

solving the following equations
=0

and

=0

and hence optimal inventory cost is obtained using the

optimal values as , and .

VIII. Numerical Example

The following randomly chosen data in appropriate units
has been used to find the optimal solution and validate the
model of the three players the producer, the distributor
and the retailer. The data are given as a=500, C=1000,
W=2000, b=100, hw=30, hr=40, =0.035, =0.040, SC=15,
k=1.3, =29, dw=20, dr=10, Ac=300 and d=30.
Using the same value of parameters as given in numerical
example we obtain the total relevant inventory costs of the
model for two cases as given in table-1.From the table we
observe that model with instantaneous deterioration and
partially backlogged in case-1 where holding cost vary in
RW after a certain time is most expensive state of affairs.
So for the case-2 where the holding cost does not vary
with the cycle length, the model is the most flexible model
and it corresponds to the least expensive circumstances.
The convexity of the model for two different cases has
been depicted in Figure-2 and Figure-3 respectively which
shows that the model has minimum value somewhere in
the range of values of parameters selected randomly.

Volume 4, Issue 5(2), September October 2015

Figure-4 Convexity of the Model for the case -2 is

represented graphically with optimal values of
and
*.
Since the model is integer in nature, reaching an
analytical solution (if any) to the problem is difficult (Gen
and Cheng, 1997). So we need to use Meta heuristic
algorithms. To solve the models under meta-heuristic
approach, four hybrid intelligent algorithms of harmony
search (Taleizadeh et al., 2009), simulated annealing
(Taleizadeh, 2008, 2009), genetic algorithm Taleizadeh
(2008, 2009), and particle swarm optimization
(Taleizadeh et al., 2009) are used. A comparison of the
results in Tables 2 and 3 for increasing and decreasing
demand show that HS method performs the best.
From Tables 2 and 3, for increasing and decreasing
demand functions, in term of purpose function's values HS
performs better than other algorithms do. Similarly, in
other examples, particle swarm optimization method has a
better solution than genetic algorithm and simulated

Page 45

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
Table: - 2 Best results of purpose function by different
algorithms for increasing demand

Table: - 3 Best results of purpose function by different

algorithms for decreasing demand.

IX. Conclusions
In this paper, we have proposed a deterministic two-ware
house inventory model for deteriorating items time
varying demand and holding cost varying with respect to
ordering cycle length with the objective of minimizing
the total inventory cost. Shortages are allowed and
partially backlogged. Two different cases have been
discussed, one with variable holding cost during the cycle
period and other with constant holding cost during total
cycle length and it is observed that during variable
holding cost the total inventory cost is more than the other
case. Furthermore the proposed model is very useful for
deteriorating items. This model can be further extended by
incorporation with other deterioration rate probabilistic
demand pattern. To solve the models, four kinds of metaheuristic algorithms or meta-heuristic hybrid algorithms:
harmony search, particle swarm optimization, genetic
algorithms, simulated annealing. Computational results
showed that HS, hybrid of HS and hybrid method of HS
performed the best, based on objective function values
respect to other kind of the algorithms. Other applications
of HS in inventory and supply chain managements can be
extended to consider pricing problems.

References
[1]. A.A. Taleizadeh, H. Shavandi, S.Aslani (2009).
Developing a MultiproductsNewsboy Problem with a
Hybrid Method of Fuzzy Simulation and Simulated
Annealing, Proceeding of 39th International
Conference
on
Computers
and
Industrial
Engineering, Paris, France, 1805-1805.
[2]. A.A. Taleizadeh, M.B. Aryanezhad, S.T.A. Niaki
(2008). Optimizing Multi-Product Multi-Constraint
Inventory Control Systems with Stochastic
Replenishment. J. Appl. Sci., 8: 1228-1234.
[3]. A.A. Taleizadeh, Moghadasi H, S.T.A. Niaki, A.
Eftekhari (2008). An EOQ Joint Replenishment
Policy to Supply Expensive Imported Raw Materials

Volume 4, Issue 5(2), September October 2015

with Payment in Advance. J. Appl. Sci., 8: 42634273.

[4]. A.A. Taleizadeh, Niaki AA, N. Shafii R. M.
Ghavamizadeh (2009). A Particle Swarm
Optimization Approach for Constraint Joint Single
Buyer-Single Vendor Inventory Problem with
Changeable Lead-Time and (r,Q) Policy in Supply
Chain. Int. J. Adv. Manuf. Syst., 51(9-12): 12091223.
[5]. A.A. Taleizadeh, S.T. Niaki, M.B. Aryanezhad
(2009). A Hybrid Method of Pareto, TOPSIS and
Genetic Algorithm to Optimize Multi-Product Multi6Constraint Inventory Control Systems with Random
Fuzzy Replenishments. Math. Comput. Model., 49:
1044-1057.
[6]. A.A. Taleizadeh, S.T. Niaki, S.V. Hosseini (2008).
Multi products, Multi Constraints Newsboy Problem
with Batch Order and Discount. Asian J. Appl. Sci.,
1: 110-122.
[7]. A.A. Taleizadeh, S.T. Niaki, S.V. Hosseini (2009).
Optimizing Multi Product Multi Constraints Biobjective Newsboy Problem with discount by Hybrid
Method of Goal Programming and Genetic
Algorithm. Eng. Optimiz., 4: 437-457.
[8]. A.A. Taleizadeh, S.T.A. Niaki (2009). A Hybrid
Method to Optimize Bi-Objective Single Period
Newsboy Problem with Fuzzy Cost and Incremental
Discount. J. Ind. Eng., Qazvin Islamic Azad
University, 3(3): 1-14.
[9]. A.A. Taleizadeh, S.T.A. Niaki, M.B. Aryanezhad
(2009). Multi-Product Multi-Constraint Inventory
Control Systems with Stochastic Replenishment and
Discount under Fuzzy Purchasing Price and Holding
Costs. Am.J. Appl. Sci., 6(1): 1-12.
[10]. A.A. Taleizadeh, S.T.A. Niaki, M.B. Aryanezhad
(2009a). Multi-product multi-constraint inventory
control systems with stochastic period length and total
discount under fuzzy purchasing price and holding
costs. Int. J. Syst. Sci., 41(10): 1187-1200.
[11]. A.A. Taleizadeh, S.T.A. Niaki, M.H. Seyed Javadi
(2009). a harmony search algorithm to optimize
chance constraint stochastic inventory systems with
dynamic demand. J. Manuf. Syst., In Press.
[12]. A.A. Taleizadeh, S.T.A. Niaki, S.V. Hosseini
(2009). Optimizing multi-product multi-constraint biobjective newsboy problem with discount by hybrid
method of goal programming and genetic algorithm.
Eng. Optimiz., 41: 437-457.
[13]. A.K. Bhunia, M. Maiti (1998). A two-warehouse
inventory model for deteriorating items with a linear
Page 46

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
trend indemand and shortages, Journal of the
Operational Research Society 49, 287-292.
[14]. A.K. Maiti, A.K. Bhunia, M. Maiti (2006). An
application of real-coded genetic algorithm (RCGA)
for mixed integer non-linear programming in twostorage multi-iteminventory model with discount
policy
Applied
Mathematics
and
Computation, Volume 183, Issue 2, Pages 903-915.
[15]. Ajay Singh Yadav, Anupam Swami (2013). A TwoWarehouse Inventory Model for Decaying Items with
Exponential Demand and Variable Holding Cost,
International Journal of Inventive Engineering and
Sciences (IJIES) ISSN: 23199598, Volume-1, Issue5.
[16]. Ali Diabat (2014) Hybrid algorithm for a vendor
managed inventory system in a two-echelon supply
chain
European
Journal
of
Operational
Research, Volume 238, Issue 1, Pages 114-121.
[17]. Ata Allah Taleizadeh, Seyed Taghi Akhavan
Niaki, Farnaz Barzinpour (2011). Multiple-buyer
multiple-vendor
multi-product
multi-constraint
supply chain problem with stochastic demand and
variable lead-time: A harmony search algorithm
Applied Mathematics and Computation, Volume 217,
Issue 22, Pages 9234-9253.
[18]. H. Altay Guvenir, Erdal Erel (1998).
Multicriteria inventory classification using a genetic
algorithm European Journal of Operational
Research, Volume 105, Issue 1, Pages 29-37.
[19]. Haixin Zhang, Yong Deng, Felix T.S. Chan,
Xiaoge Zhang (2013). A modified multi-criterion
optimization genetic algorithm for order distribution
in collaborative supply chain Applied Mathematical
Modelling, Volume 37, Issues 1415, Pages 78557864.
[20]. Hertely V. Ronald. (1976). On the EOQ model two
levels of storage. Opsearch, 13, 190-196.
[21]. K. D. Rathod and P. H. Bhathawala (2013).
Inventory model with inventory-level dependent
demand rate, variable holding cost and shortages,
International Journal of Scientific & Engineering
Research, Volume 4, Issue 8, August-2013.
[22]. K.L. Mak, Y.S. Wong, G.Q. Huang (1999).
Optimal inventory control of lumpy demand items
using genetic algorithms Computers & Industrial
Engineering, Volume 37, Issues 12, Pages 273-276.
[23]. K.V.S. Sarma (1987). A deterministic order level
inventory model for deteriorating items with two
storage facilities. European Journal of Operational
Research 29,70-73.

Volume 4, Issue 5(2), September October 2015

[24]. K.V.S. Sarrna (1983). A deterministic inventory

model with two level of storage and an optimum
release rule,Opsearch 20, 175-180.
[25]. Keyvan Sarrafha, Seyed Habib A. Rahmati, Seyed
Taghi Akhavan Niaki, Arash Zaretalab (2015). A
bi-objective integrated procurement, production, and
distribution problem of a multi-echelon supply chain
network design: A new tuned MOEA Computers &
Operations Research, Volume 54, Pages 35-51.
[26]. Maya Gyan, A.K. Pal (2009). A two warehouse
Inventory model for deteriorating items with stock
dependent demand rate and holding cost,Oper. Res.
Int. J.vol(9),153-165.
[27]. N. Jawahar, A.N. Balaji (2009). A genetic
algorithm for the two-stage supply chain distribution
problem associated with a fixed charge European
Journal of Operational Research, Volume 194, Issue
2, Pages 496-537.
[28]. Pavan Kumar Naraharisetti, Iftekhar A. Karimi,
Rajagopalan Srinivasan (2007). Optimal supply
chain redesign using genetic algorithm Computer
Aided Chemical Engineering, Volume 24, Pages 703708.
[29]. R.P. Tripathi (2013). Inventory model with different
demand rate and different holding cost, International
Journal of Industrial Engineering Computations 4
(2013) 437446.
[30]. S. Kar. A.K. Bhunia, M. Maiti (2001).
Deterministic inventory model with two levels of
storage, a linear trendin demand and a fixed time
horizon. Computers & Operations Research 28 ,
1315-1331.
[31]. S.A. Torabi, S.M.T. Fatemi Ghomi, B. Karimi
(2006). A hybrid genetic algorithm for the finite
horizon economic lot and delivery scheduling in
supply chains European Journal of Operational
Research, Volume 173, Issue 1, Pages 173-189.
[32]. S.M. Disney, M.M. Naim, D.R. Towill (2000).
Genetic
algorithm optimisation
of
a
class
of inventory control systems International Journal of
Production Economics, Volume 68, Issue 3, Pages
259-278
[33]. S.P. Nachiappan, N. Jawahar (2007). A genetic
algorithm for optimal operating parameters of VMI
system in a two-echelon supply chain European
Journal of Operational Research, Volume 182, Issue
3, Pages 1433-1452.
[34]. Sung-Pil Hong, Yong-Hyuk Kim (2009). A genetic
algorithm for joint replenishment based on the
exact inventory cost Computers & Operations
Research, Volume 36, Issue 1, Pages 167-175.
Page 47

International Journal of Emerging Trends & Technology in Computer Science (IJETTCS)

Web Site: www.ijettcs.org Email: editor@ijettcs.org
Volume 4, Issue 5(2), September - October 2015
ISSN 2278-6856
[35]. T.A. Murdeshwar, Y.S. Sathe (1985). Some aspects
of lot size model with two levels of storage, Opsearch
22,255-262.
[36]. T.P.M. Pakkala, K.K. Achary (1992). A
deterministic inventory model for deteriorating items
with two ware-houses and finite replenishment rate,
European Journal of Operational Research 57, 157167.
[37]. Vinod Kumar Mishra (2013). An inventory model
for deteriorating items withtime-dependent demand
and time-varying holding cost under partial
backlogging, Journal of Industrial Engineering
International 2013, 9:4.
[38]. Wei-Chang Yeh, Mei-Chi Chuang (2011). Using
multi-objective genetic algorithm for partner selection
in green supply chain problems Expert Systems with
Applications, Volume 38, Issue 4, Pages 4244-4253.
[39]. Z.H. Che, C.J. Chiang (2010). A modified Pareto
genetic algorithm for multi-objective build-to-order
supply chain planning with product assembly
Advances in Engineering Software, Volume 41,
Issues 78, Pages 1011-1022.

Volume 4, Issue 5(2), September October 2015

Page 48