(IJCSIS) International Journal of Computer Science and Information Security,

Vol. 7, No. 1, 2010

Multi-Product Inventory Optimization using

Uniform Crossover Genetic Algorithm
[Link] [Link]
Assistant Professor Professor and Head
Department of Computer Science and Engineering Department of Mechanical Engineering
Park College of Engineering and Tekhnology Coimbatore Institute of Technology
Coimbatore – 641659, Tamilnadu, India Coimbatore – 641014, Tamilnadu, India

[Link]
Research Scholar
Department of Computer Science and Engineering
Anna University – Coimbatore, Tamilnadu, India

of demand, while striking a reasonable balance between the

Abstract - Inventory management is considered to be an cost of holding the items in inventory and the penalty (loss of
important field in Supply Chain Management because the cost of sales and goodwill, say) of running out. The item may be a
inventories in a supply chain accounts for about 30% of the value commodity sold by a store; it may be spare machine parts in a
of the product. The service provided to the customer eventually
factory; it may be railway wagons; it may be cash in the bank
gets enhanced once the efficient and effective management of
inventory is carried out all through the supply chain. The precise
to meet the customers’ demand. It is indeed surprising to find
estimation of optimal inventory is essential since shortage of that a very wide variety of seemingly different problems can
inventory yields to lost sales, while excess of inventory may result be mathematically formulated as an inventory-control
in pointless storage costs. Thus the determination of the problem. There are, of course, several different models of
inventory to be held at various levels in a supply chain becomes inventory systems. There are three types of expenses
inevitable so as to ensure minimal cost for the supply chain. The associated with inventory systems. The relative importance of
minimization of the total supply chain cost can only be achieved these will depend on the specific system. They are: (i)
when optimization of the base stock level is carried out at each administrative cost of placing an order, called reorder cost or
member of the supply chain. This paper deals with the problem set cost; (ii) cost of maintaining an inventory, called inventory
of determination of base-stock levels in a ten member serial holding cost a carrying cost, which includes storage charge,
supply chain with multiple products produced by factories using interest, insurance, etc., a (iii) shortage cost is a loss of profit,
Uniform Crossover Genetic Algorithms. The complexity of the goodwill, etc., when run out of stock. All the above should be
problem increases when more distribution centers and agents optimized for efficient supply chain management.
and multiple products were involved. These considerations
leading to very complex inventory management process has been A. Inventory Control in Supply Chain Management
resolved in this work. It has been stated by several people that the focus point of
supply chain management is inventories and inventory
Keywords: Supply Chain Management, Inventory control. To transfer their focus from scheming logistical costs
Optimization, Base Stock, Uniform Crossover, Genetic to investigate supply chains [2] few food manufacturers and
Algorithm (GA), Supply Chain Cost grocers formed Efficient Consumer Response in the year
I. INTRODUCTION 1992. The major competitive factor for companies focused on
value creation for end consumers is the customer service. In
Supply Chain Management (SCM) is an efficient general, firms hold inventory for two major reasons, to lessen
management of the complete end to end process, starting from costs and to improve customer service. The inspiration for
the design of the product or service to the time when it has each varies as firms stabilize the problem of having too much
been sold, consumed and finally gotten rid of by the inventory (which can direct to high costs) versus having very
consumer. This complete process includes product design, small inventory (which can direct to lost sales) [3].
procurement, planning and forecasting, production,
distribution, fulfillment and after sales supports. A company’s Supply chain management leads to cost savings, mainly in
competitiveness in the global economy can be increased only the course of lessening in inventory. Inventory costs have got
with the aid of effective SCM. This involves complex reduced by about 60% from 1982, whereas transportation
strategic, tactical and operational decisions that often require costs have fallen by 20% [4]. These cost savings have led
an in-depth understanding of industry-specific issues, which many people to follow inventory-reduction strategies in the
ranges from network design to production sourcing and from supply chain. To deal with inventory, firms make use of one
production planning and inventory management to scheduling of three common approaches. First of all, the majority of
[1]. retailers make use of an inventory control approach,
monitoring inventory levels by item. The second thing is,
The inventory management problem is one of maintaining manufacturers are typically more concerned with production
an adequate supply of some item to meet an expected pattern scheduling and use flow management to deal with inventories.

170 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
Third, numerous firms (for the majority part those handling meeting market demand [14-16]. The design and management
raw materials or in extractive industries) do not keenly deal of the storage policies and procedures for raw materials, work-
with inventory [5]. in-process inventories, and typically, final products are
illustrated by the inventory control. The costs and lead times
The inventory management is influenced by the nature of can be reduced and the responsiveness to the changing
demand, depending on whether demand is derived or customer demands can be significantly improved and
independent. Independent demand comes up from demand for subsequently inventory can be optimized by the effective
an end product. End products are found all through the supply handling of the supply chain [17].
chain. By definition, a self-governing demand is uncertain,
meaning that extra units or safety stock must be accepted to There are several reasons for manufacturers’ increasing
guard against stock outs. While managing uncertainty, the focus on optimizing inventory by applying the latest tools and
objective should be to minimize the inventory levels and also techniques for inventory control. Traditionally, competitive
meet customer expectation. Supply chain coordination can pressure has always driven manufacturers to seek enhanced
reduce the ambiguity of intermediate product demand, in that capabilities to reduce inventory levels; to enhance service
way reducing inventory costs [3, 6]. levels and supply availability; and to establish the right
product inventory mix and level in each geography and
Since Ford Harris’ renowned Economic Order Quantity channel. A key driver of the renewed focus on inventory lies
(EOQ) model was first proposed in 1913, the inventory in the recognition that traditional techniques are failing to
control has been rewarded immense awareness for a long time reign in inventories in the wake of increased supply chain
because of its significance in the cost control. To lessen the complexity. This complexity is characterized by increased
total expected inventory costs per unit time while satisfying uncertainty. Demand is more volatile and therefore less
the customer demand on time [7] is one of the major predictable. This is true not only for aggregate demand but for
objectives. Inventory control for large-scale supply chains is forecasting splits and volumes across channels and markets.
well recognized [8-10] as an essential problem with several Traditionally three strategies have been employed by
applications together with manufacturing systems, logistics manufacturers to address uncertainty; a) increase inventory
systems, communication networks, and transportation systems levels to hedge against uncertainty; b) develop supply chain
[11]. It is essential to locate the apt mechanism for flexibility to be more responsive to uncertainty; c) improve
coordinating the inventory processes that are controlled by forecast accuracy so that less uncertainty propagates to the
independent partners, in order to find out the right ordering manufacturing floor. Inventory optimization techniques and
quantity and inventory level amid partners in the chain. For technologies map to the flexibility and accuracy strategies.
example, the manufacturer make use of the periodic review [18].
and lot sizing policy to manage its inventory and the retailer
employs the periodic review with target stock level to control Inventory Optimization characterizes the supply network
its inventory and more [12]. uncertainty present in a variety of specific steps or links in
manufacturing and distribution processes. Advanced
B. Inventory Optimization in Supply Chain Management mathematical models are then solved to identify optimal
The effective management of the supply chain has become inventory policies, stocking locations, or quantities.
unavoidable these days due to high expectation in customer The uncertainty addressed by IO include: demand uncertainty,
service levels [13]. The supply chain cost was immensely cycle time variability and replenishment lead time
influenced by the overload or shortage of inventories. Thus variability.[18] Efficient management of the supply chain, i.e.
inventory optimization has transpired into one of the most the reduction of the costs and lead times and vastly enhanced
important topics as far as supply chain management is responsiveness to the changing customer demands lead to an
considered [14-16]. optimized inventory.
To exploit economies of scale and order in large lots, the
important issues in supply chain is to optimize the inventory II. RELATED WORKS - REVIEW
level by considering various costs in maintaining a high
A. Review of Base-Stock based Inventory Control Models
service level towards the customer. Since, the cost of capital
tied up in inventory is more, the inventory decision in the The inventory control problem for a single class assembly
supply chain should be coordinated without disturbing the network which operates under a modified echelon base-stock
service level. The coordination of inventory decision within policy was studied by Vecchio and Paschalidis [19].
an entity is viable, but not between the entities. So the An approach to find close-to-optimal echelon stock levels that
integration of the entities to centralize the inventory control is minimize inventory costs while guaranteeing stockout
needed. probabilities to stay below some predefined levels was
developed by them. They reduced the safety stock selection to
Inventory Optimization [IO] application organizes the a deterministic nonlinear optimization problem on the basis of
latest techniques and technologies, thereby assisting the the large deviations techniques. In addition, they analyzed as
improved inventory visibility, the enhancement of inventory to how a supplier can interact with a buyer to reach a mutually
control and its management across an extended supply beneficial mode of operations, using their inventory control
network. Some of the design objectives of inventory approach. Their interaction takes the form of a supply contract
optimization are to optimize inventory strategies, thereby by which explicit QoS guarantees is enforced.
enhancing customer service, reducing lead times and costs and The applications in a distributed fashion with neither the

171 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
supplier nor the buyer revealing their corresponding cost The inventory levels across supply chain members were
structures applying the joint optimization algorithm was obtained with the aid of a search routine.
proposed by the authors.
B. Review of Optimization based Inventory Control Models
A model of supply chain consisting of n production Sukran Kadipasaoglu et al. [1] provided a study on the
facilities in tandem and producing a single product class was market characteristics and competitive priorities,
considered by Ioannis CH. Paschalidis et al. [20]. The finished manufacturing environment, logistics and distribution
goods inventory maintained in front of the most downstream activities, and supply chain planning and control activities for
facility is used to meet the external demand while backlogging polymer manufacturers. They have described polymer
of unsatisfied demand was performed. The facility at stage 1 distribution network optimization, production/distribution
produced if inventory has fallen below a certain level wi and planning, production scheduling, demand management,
idles otherwise on the basis of a base-stock production policy available-to-promise, and inventory planning activities
adopted at each stage of the supply chain. In order to pertaining to supply chain planning and control. Besides, they
minimize expected inventory costs at all stages subject to illustrated the applications existing in a commercial DSS that
maintaining the stock out probability at stage1 below a support these activities. They have as well described about
prescribed level, they necessitated the optimization of the diverse issues that continue to confront supply chain managers
hedging vector W= (w1,...,wn). They made assumptions on in polymer manufacturing. It encompasses forecasting for the
demand and production processes that included auto huge number of product–customer combinations,
correlated stochastic processes, which were relatively general identification of safety stock requirements, administering
modeling. They have combined analytical (Large derivations) production schedule changes, business process management
and sample path based (perturbation analysis) techniques to throughout DSS implementation and data mapping for
solve the stochastic optimization problem. The existence of a decision support. Their research contributes to the supply
natural synergy between those two approaches has been chain literature by proffering a suitable context for
demonstrated. investigating supply chain-related issues. Through discussion
An attempt was made to optimize the inventory (i.e. base- and characterization of the polymer supply chain, they
stock) levels of a single product at different members in a recognized the specific issues of concern to potential
serial supply chain with the objective of minimizing the Total researchers and to supply chain professionals.
Supply Chain Cost (TSCC), by Sudhir Ryan Daniel and The effect of product variety on supply-chain
Chandrasekharan Rajendran [21] and P. Radhakrishnan et al. performance, which is measured in terms of expected lead
[22]. The performance measure considered, which is a good time and expected cost at the retailers, was analyzed by Ulrich
representation of the system-wide total cost is the TSCC. W. Thonemann and James R. Bradley [25]. They took a
In order to optimize the base-stock levels, a genetic algorithm supply chain with a single manufacturer and multiple retailers
(GA) has been proposed. To analyze the performance of the into account. If setup times are significant, the effect of
supply chain (operating with deterministic and stochastic product variety on cost where the cost increases proportionally
replenishment lead times) for the base-stock levels that are to the square root of product variety is substantially greater
generated by the proposed GA and other solution procedures than that suggested by the risk-pooling literature for perfectly
considered in this study, different supply chain settings are flexible manufacturing processes. An illustration that
simulated. They demonstrated that their proposed GA required underestimates the cost of product variety, leads companies to
significantly less computing effort to perform very well in offer product variety that is greater than optimal was made as
terms of yielding solutions that are not significantly different well. In conclusion, they illustrated that by reducing the setup
from the optimal solutions (obtained through complete time, the unit manufacturing time, the number of retailers, or
enumeration of solution space). the demand rate the supply-chain performance can be
A beneficial industry case applying Genetic Algorithms managed. The fact that complex mathematical approaches are
(GA) has been proposed by [Link] and [Link] [23]. often not applied in practice was recognized by the authors.
The case has made use of GAs for the optimization of the total Nevertheless, practitioners who used the simple models to
cost of multiple sourcing supply chain system. The system has estimate the effect of their decisions often appreciated these
been exemplified by a multiple sourcing model with stochastic models.
demand. A mathematical model has been implemented to The inventory and supply chain managers are mainly
portray the stochastic inventory with the many to many concerned holding of the excess stock levels and hence the
demand and transportation parameters as well as price increase in the holding cost. Meanwhile, there is possibility
uncertainty factors A genetic algorithm which has been for the shortage of products. For the shortage of each product
approved by Lo [24] deals with the production-inventory there will be a shortage cost. Holding excess stock levels as
problem with backlog in the real situations, with time-varied well as the occurrence of shortage for products lead to the
demand and imperfect production due to the defects in increase in the supply chain cost. The factory may
production disruption with exponential distribution. Besides manufacture any number of products, each supply chain
optimizing the number of production cycles to generate a (R, member may consume a few or all the products and each
Q) inventory policy, an aggregative production plan can also product is manufactured using a number of raw materials
be produced to minimize the total inventory cost on the basis sourced from many suppliers. All these factors pose additional
of reproduction interval searching in a given time horizon.

172 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
challenge in extracting the exact product and the stock levels The robustness of the Genetic algorithms as search
that influence the supply chain cost heavily. techniques have been theoretically and empirically proved
[27]. The artificial individual is the basic element of a GA. An
Many well-known algorithmic advances in optimization artificial individual consists of a chromosome and a fitness
have been made, but it turns out that most have not had the value, similar to a natural individual. The individual's
expected impact on the decisions for designing and optimizing likelihood for survival and mating is determined by the fitness
supply chain related problems. For example, some function [28]. In accordance with the Darwin’s principle,
optimization techniques are of little use because they are not individuals superior to their competitors, are more likely to
well suited to solve complex real logistics problems in the promote their genes to the next generations. In accordance
short time needed to make decisions. Also some techniques with this concept, in Genetic Algorithms, a set of encoded
are highly problem-dependent and need high expertise. This parameters are mapped into a potential solution, named
adds difficulties in the implementations of the decision chromosome, to the optimization problem [29].
support systems which contradicts the tendency to fast The population of candidate solutions is obtained through the
implementation in a rapidly changing world. IO techniques process of selection, recombination, and mutation performed
need to determine a globally optimal placement of inventory, in an iterative manner. [30].
considering its cost at each stage in the supply chain and all
the service level targets and replenishment lead times that Chromosomes refer to the random population of encoded
constraint each inventory location about the estimation of the candidate solutions with which the Genetic algorithms initiate
exact amount of inventory at each point in the supply chain with. [27]. Then the set (called a population) of possible
free of excesses and shortages although the total supply chain solutions (called chromosomes) are generated [31]. A function
cost is minimized. Owing to the fact that shortage of inventory assigns a degree of fitness to each chromosome in every
yields to lost sales, whereas excess of inventory may result in generation in order to use the best individual during the
pointless storage costs, the precise estimation of optimal evolutionary process [32]. In accordance to the objective, the
inventory is indispensable [26]. In other words, there is a cost fitness function evaluates the individuals [30].
involved in manufacturing any product in the factory as well Each chromosome is evaluated using a fitness function and a
as in holding any product in the distribution center and agent fitness value is assigned. Then, three different operators-
shop. More the products manufactured or held, higher will be selection, crossover and mutation- are applied to update the
the holding cost. Along with this, low lead time results in population. A generation refers to an iteration of these three
operators [33]. The promising areas of the search space are
III. OBJECTIVES focused in the selection step. The selection process typically
keeps solutions with high fitness values in the population and
The supply chain cost can be minimized by maintaining rejects individuals of low quality [30]. Hence, this provides a
optimal stock levels in each supply chain member. There is a means for the chromosomes with better fitness to form the
necessity of determining the inventory to be held at different mating pool (MP) [31]. After the process of Selection, the
stages in a supply chain that will minimize the total supply Crossover is performed.
chain cost i.e., minimizing holding and shortage cost. The
inventory control for more number of products along with B. Uniform Crossover
different levels of supply chain is a complex task. The In the crossover operation, two new children are formed
approach aims to make use of the meta heuristic algorithms by exchanging the genetic information between two parent
like Genetic algorithm for the prediction of the optimal stock chromosomes. Multipoint crossover defines crossover points
levels to be maintained, so as to minimize the total supply as places between loci where an individual can be split.
chain inventory cost, comprising holding and shortage costs at Uniform crossover generalizes this scheme to make every
all members of the supply chain. The genetic algorithm is locus a potential crossover point. A crossover mask, the same
proposed that considers all these factors that are mentioned length as the individual structure is created at random and the
hitherto such that the analysis paves the way for minimizing parity of the bits in the mask indicate which parent will supply
the supply chain cost by maintaining optimal stock levels in the offspring with which bits. This method is identical to
each supply chain member. discrete recombination.
A. Genetic Algorithm Consider the following two individuals with 11 binary
Genetic algorithm is a randomized search methodology variables each:
having its roots in the natural selection process. Initially the Individual 1 0 1 1 1 0 0 1 1 0 1 0
neighborhood search operators (crossover and mutation) are Individual 2 1 0 1 0 1 1 0 0 1 0 1
applied to the preliminary set of solutions to acquire For each variable the parent who contributes its variable to
generation of new solutions. Solutions are chosen randomly the offspring is chosen randomly with equal probability. Here,
from the existing set of solutions where the selection the offspring 1 is produced by taking the bit from parent 1 if
probability and the solution’s objective function value are the corresponding mask bit is 1 or the bit from parent 2 if the
proportional to each other and eventually the aforesaid corresponding mask bit is 0. Offspring 2 is created using the
operators are applied on the chosen solutions. Genetic inverse of the mask, usually.
algorithms have aided in the successful implementation of
solutions for a wide variety of combinatorial problems. Sample 1 0 1 1 0 0 0 1 1 0 1 0
Sample 2 1 0 0 1 1 1 0 0 1 0 1

173 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
After crossover the new individuals are created: products that would be supplied to the distribution centers.
offspring 1 1 1 1 0 1 1 1 1 1 1 1 From the distribution center, the stocks will be moved to the
offspring 2 0 0 1 1 0 0 0 0 0 0 0 corresponding agents.
Uniform crossover has been claimed to reduce the bias To make the inventory control effective, the most primary
associated with the length of the binary representation used objective is to predict where, why and how much of the
and the particular coding for a given parameter set. This helps control is required which is made through the proposed
to overcome the bias in single-point crossover towards short GAUX methodology. The proposed methodology is aimed at
substrings without requiring precise understanding of the determining the specific product that needs to be concentrated
significance of the individual bits in the individual’s on and the amount of stock levels of the product to be
representation. How uniform crossover may be parameterized maintained by the different members of the supply chain. The
by applying a probability to the swapping of bits was methodology also analyses whether the stock level of the
demonstrated by William M. Spears et al.[34]. particular product needs to be in abundance, in order to avoid
shortage of the product or needs to be held minimal in order to
This extra parameter can be used to control the amount of minimize the holding cost.
disruption during recombination without introducing a bias
towards the length of the representation used.
The chromosome cloning takes place when a pair of
chromosomes does not crossover, thus creating off springs
that are exact copies of each parent [28].
The ultimate step in each generation is the mutation of
individuals through the alteration of parts of their genes [26].
Mutation alters a minute portion of a chromosome and thus
institutes variability into the population of the subsequent
generation [27]. Mutation, a rarity in nature, denotes the
alteration in the gene and assists us in avoiding loss of genetic
diversity [26]. Its chief intent is to ensure that the search
algorithm is not bound on a local optimum [28].
IV. INVENTORY OPTIMIZATION ANALYSIS USING UNIFORM
CROSSOVER GENETIC ALGORITHM
The proposed method uses the Genetic Algorithm with
Uniform Crossover to study the stock level that needs
essential inventory control. This is the pre-requisite
information that will make any kind of inventory control
effective. In practice, the supply chain is of length n, means
having n number of members in supply chain such as factory,
distribution centers, suppliers, retailers and so on. Fig. 2 Genetic Algorithm steps for the proposed
The exemplary supply chain taken for the implementation of inventory management analysis
the proposed method is shown in Fig. 1. The methodology as shown in Fig. 2 would analyze the
past records very effectively and thus facilitate efficient
inventory management with the aid of Genetic Algorithm.
The analysis is initiated by the selection of valid records.
The validation of records is done over the records of past
periods. The stock levels at the different supply chain
members are held in the dataset for different products, namely
P1, P2, P3, P4, P5, P6, P7, etc. Seven products have been
considered for the analysis, hence the stock levels for the
seven products at each member of the chain throughout the
past period are considered as data set as shown in the Table 1.
Fig. 1 Three Stage - 10 Member Supply Chain For the valid record set selection, records having nil values are
Fig. 1 exhibits that a factory is the parent of the chain and neglected and the records having positive or negative values
it is having three distribution centers Distribution Center 1, are selected for the analysis. This can be done by means of
Distribution Center 2 and Distribution Center 3. Each clustering algorithms, extraction algorithms or by any of the
distribution center further comprises of several agents but as data mining functions. Hence the extraction function results in
stated in the example case, each Distribution center is having data sets having either positive or negative values. The record
two agents. So, in aggregate there are six agents, Agent 1 and set having positive values represents excess stock levels and
Agent 2 for Distribution Center 1, Agent 3 and Agent 4 for the negative values represent shortage level of a particular
Distribution Center 2 and Agent 5 and Agent 6 for product at a particular member of the supply chain. Then the
Distribution Center 3. The factory manufactures different

174 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
data set is subjected to Genetic Algorithm and the various inventory values; and log [1- (Nrep / Nt)] will ensure minimum
steps performed in the genetic algorithm are discussed below. value corresponding to the maximum probability; So, the
fitness function is structured to retain the minimum value
A. Generation of Individuals corresponding to the various chromosomes being evaluated
The randomly generated initial chromosome is created by iteration after iteration and this in turn ensures that the fitness
having the stock levels within the lower limit and the upper function evolution is towards optimization.
limit for all the contributors of the supply chain, factory and
the distribution centers. As known, chromosome is constituted C. Selection
by genes which defines the length of the chromosomes. The selection operation is the initial genetic operation
The stock level of each member of the chromosome is referred which is responsible for the selection of the fittest
as gene of the chromosome. Hence for n length supply chain, chromosome for further genetic operations. The fitness
the chromosome length is also n . Since a 10 member supply function is carried out for each chromosome and the
chain is used for illustration, the length of the chromosome n chromosomes are sorted on the basis of the result of the
is 10, i.e. 10 genes. And the chromosome representation is fitness function and ranked. The chromosome generating
pictured in Fig. 3. Each gene of the chromosome is value as minimum as possible will be selected by the fitness
representing the amount of stock that is in excess or in function and will be subjected further to the genetic
shortage at the respective members of the supply chain. operations, crossover and mutation.
D. Uniform Crossover
Among the numerous crossover operators in practice, a
uniform crossover is chosen in this proposed method for its
advantages over the other forms. Uniform crossover is global
and less biased when compared to that of standard and one
Fig. 3 Random individual generated for the genetic operation point crossover. Uniform crossover does not select a set of
These kinds of chromosomes are generated for the genetic crossover points. It simply considers each bit position of the
operation. Initially, only two chromosomes will be generated two parents, and swaps the two bits with a probability of 50%.
and from the next generation a single random chromosome With large search spaces, a GA using uniform crossover
value will be generated. The chromosomes thus generated is outperforms a GA using one point crossover, which in turn
then applied to find its number of occurrences in the database outperforms a GA using two point crossover [35-36]. From
content by using a Select count ( ) function. The function will the matting pool, two chromosomes are subjected for the
give the number of occurrences/ repetitions of the particular uniform crossover. The chromosomes initially selected and
amount of stock level for the ten members Nrep that are going after undergoing uniform crossover operation performed in
to be used further in the fitness function. this analysis is pictured in Fig. 4. As soon as the crossover
operation is completed, the genes of the two chromosomes
B. Evaluation of Fitness function present get interchanged.
A specific kind of objective function that enumerates the Before Crossover
optimality of a solution in a genetic algorithm in order to rank
certain chromosome against all the other chromosomes is
known as Fitness function. Optimal chromosomes, or at least
chromosomes which are near optimal, are permitted to breed
and merge their datasets through one of the several techniques
available in order to produce a new generation that will be
better than the ones considered so far. After Crossover
The fitness function is given by:

 N rep 
f (k )  log1  , k  1, 2 ,3   , m
 Nt  (1)
Fig. 4 Chromosomes after uniform crossover operation
where,
N rep is the number of repetitions of records of similar E. Mutation
stock levels that occurs throughout the period; The crossover operation is succeeded by the final stage of
genetic operation known as Mutation. In the mutation, a new
N t is the total number of records of inventory values chromosome is obtained. This chromosome is totally new
obtained after clustering; from the parent chromosome. The concept behind this is the
m is the total number of chromosomes for which the child chromosome thus obtained will be fitter than the parent
fitness function is calculated. chromosome. The performance of mutation operation is
shown in Fig. 5.
In the fitness function, the ratio (Nrep / Nt) plays the role of
finding the probability of occurrence of a particular record of

175 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
V. EXPERIMENTAL RESULTS
The approach suggested for the optimization of inventory
level and thereby an efficient supply chain management has
been implemented in the platform of LabVIEW 2009.
The database consists of the records of stock levels held by
each member of the supply chain for every period. For
implementation, seven different products in circulation with
the ten member supply chain network have been considered.
A sample set of data from a large database used in the
implementation is given in Table 1.
Fig. 5 Chromosome subjected to mutation operation
Table 1. A Sample Dataset Constituted by the Product Identification
along with its Stock Levels in Each Member of the Supply Chain
Four mutation points are chosen as shown in Fig. 5.
The mutation is done on the particular gene present at the PI F1 DC1 DC2 DC3 A1 A2 A3 A4 A5 A6
Mutation points. This pointing of gene is done randomly. 7 -371 -736 -299 634 448 756 340 -736 -778 863
5 -407 379 -981 -864 -391 999 -196 307 -171 -529
Hence, the four mutation points may point any of the ten 2 -146 -604 443 746 -561 -734 445 424 -891 -824
genes. 4 -962 -524 -685 -254 205 446 -469 108 346 840
The process explained so far will be repeated along with 3 -834 266 969 965 735 244 -752 133 -554 -939
3 -449 -282 577 -926 -414 -200 -743 850 196 851
the new chromosome obtained from the previous process. 4 540 -830 -835 882 -379 768 -635 -112 539 107
In other words, at the end of each of the iteration, a best 3 -778 -313 629 -690 824 -927 850 307 -171 -529
chromosome will be obtained. This will be included with the 2 351 293 328 -732 357 -566 685 424 -891 -824
newly generated random chromosome for the next iteration. 1 500 108 490 -345 -236 108 -931 -260 -144 162
When the number of iterations is increased then the obtained 5 844 -728 286 740 686 -421 424 -792 -927 -879
solution moves very closer to the accurate solution. More the 4 -321 902 -450 -260 -144 162 238 307 -171 -529
number of iterations the more accurate the optimal solution 3 775 -394 -520 -792 -927 -879 -507 424 -891 -824
4 794 932 -584 307 -171 -529 -503 108 346 840
will be. Eventually with the help of the Genetic algorithm, the
2 -122 -686 -620 424 -891 -824 941 133 -554 -939
best stock level to be maintained in the members of the supply 6 235 464 401 108 346 840 -934 464 401 108
chain could be predicted from the past records, so that the loss 5 218 -848 836 133 -554 -939 -834 -848 836 133
due to the holding of excess stock level and shortage level can 4 489 409 148 850 196 851 -495 -144 162 238
be reduced leading to an optimal inventory solution. 3 -422 638 676 -112 539 107 -440 -927 -879 -507
5 893 520 -423 -736 -778 863 -335 676 -112 539

Fig. 6 Inventory Optimization Tool created in LabVIEW 2009

176 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
In the sample database tabulated in Table 1, the first field sufficient to reduce the loss either due to the holding of excess
comprises of the product Identification (PI) and the other stocks or due to the shortage of stocks. By focusing on the
fields are related with the stock levels that were held by the excess/shortage inventory levels and initiating appropriate
respective ten members of the supply chain network. For steps to eliminate the same at each member of the chain, it is
example, the first attribute and first field of the database is ‘7’ possible to optimize the inventory levels in the upcoming
which refers the Product I.D. ‘7’. The corresponding fields of period and thus minimize the supply chain cost.
the same attribute denote the stock levels of the product I.D.
‘7’ in the respective members of the supply chain. Similarly, The organization should take necessary steps to decrease
different sets of stock levels for different products are held by the production of product 6 in the factory by 1800 units to
the database. make up for the predicted excess; increase the inventory level
of product 1 by 300 units in distribution center 1 to make up
As per the proposed analysis based on Uniform Crossover for the predicted shortage, reduce inventory level of product 1
GA, two random initial chromosomes are generated as shown by 800 units and 350 units in distribution centers 2 and 3
in Fig. 3. These initial chromosomes are subjected for the respectively to make up for the predicted excess.
genetic operators, Uniform Crossover and Mutation.
The resultant chromosome thus obtained after the application Agent 1 should decrease the inventory level of product 6
of crossover and mutation is shown in Fig. 5. by 650 units. Agent 2 should decrease the inventory level of
product 6 by 700 units. Agent 3 and Agent 4 should increase
Fig. 6 shows the main window of the tool created for the inventory level of product 6 by 800 units and 200 units
inventory optimization using genetic algorithm in LabVIEW respectively to make up for the predicted excess / shortage.
2009. The tool created is based on the uniform crossover The inventory level of product 6 should be decreased by 100
genetic algorithm. units and 225 units by Agent 5 and Agent 6 respectively.
Thus by following the predicted stock levels, the
The window displays the fitness function used, uniform excess/shortage inventory levels can be avoided in the
crossover sample bit pattern randomly generated with upcoming period and thus the increase of supply chain cost
probability of 0.5, the points at which mutation should occur, can also be avoided. The analysis extracts an inventory level
the end iteration count upon which the fitness function that made a remarkable contribution towards the increase of
stabilizes, no. of supply chain members and population size. supply chain cost, and in turn enabled to predict the future
The best chromosome obtained after the required iterations optimal inventory levels to be maintained in all the supply
and the fitness value and plot generated for the iteration value chain members with the aid of these levels. Therefore it is
are also displayed. possible to minimize the supply chain cost by maintaining the
optimal stock levels that was predicted from the inventory
The organization can decide about the quantum of analysis, and thus making the inventory management more
iterations for running the simulation to arrive at the optimal effective and efficient.
solution. As long as minimization of the fitness function is
still possible, then the iteration continues till such a time that VII. CONCLUSION
no improvement in the fitness function value is noticeable.
Inventory management is an important component of
After a certain number of iterations, if the fitness function
supply chain management. An innovative and efficient
value is not improving from the previous iterations, then this
methodology that uses Genetic Algorithms with Uniform
is an indication that the fitness function value is stabilizing
Crossover to precisely determine the most probable excess
and the algorithm has converged towards optimal solution.
stock level and shortage level required for inventory
This inference is useful for deciding the number of iterations
optimization in the supply chain such that the total supply
for running the GA simulation as well as this may be used as
chain cost is minimal is proposed using LabVIEW 2009.
the stopping criteria for the algorithm. For greater accuracy,
the number of iterations should be sufficiently increased and The optimized stock level at all members of the supply
run on the most frequently updated large database of past chain is obtained by following the proposed genetic algorithm.
records. Thus the proposed work gives a better prediction of stock
levels amid diverse stock levels at all members of the supply
As for our iteration value of ‘200’, the resultant
chain. The complexity of increasing the number of products
chromosome moved towards the best chromosome after each
through the supply chain has been resolved by the proposed
iterative execution. Hence at the end of the execution of 200th
approach. Products due to which the members of the supply
iteration, best chromosome obtained is shown in Fig. 7.
chain incur extra holding or shortage cost are also determined.
6 1800 -300 800 350 650 700 -800 -200 100 225 More specifically, the inventory is optimized in the whole
supply chain regardless of the number of products and the
Fig. 7 The final best chromosome obtained after 200th iteration number of members in the supply chain. The proposed
approach of inventory management has achieved the
VI. DISCUSSION OF RESULTS objectives which are the minimization of total supply chain
The final chromosome obtained from the GA based cost and the determination of the products due to which the
analysis shown in the Fig. 7 is the inventory level that has the supplier endured either additional holding cost or shortage
potential to cause maximum increase of supply chain cost. cost.
It is inferred that controlling this resultant chromosome is

177 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
REFERENCES [23]. K. Wang and Y. Wang, “Applying Genetic Algorithms to Optimize the
Cost of Multiple Sourcing Supply Chain Systems - An Industry Case
[1]. Sukran Kadipasaoglu, Jennifer Captain and Mark James, “Polymer Study”, Studies on Computational Intelligence, Vol. 92, pp. 355-372,
Supply Chain Management”, International Journal on Logistics Systems 2008.
and Management, Vol. 4, No. 2, pp. 233-253, 2008. [24]. Chih-Yao Lo, “Advance of Dynamic Production-Inventory Strategy for
[2]. R. King and P. Phumpiu, “Reengineering the food supply chain: The Multiple Policies Using Genetic Algorithm”, Information Technology
ECR initiative in the grocery industry”, American Journal of Journal, Vol. 7, pp. 647-653, 2008.
Agricultural Economics, Vol. 78, pp. 1181-1186, 1996. [25]. Ulrich W. Thonemann and James R. Bradley, “The effect of product
[3]. Frank Dooley, “Logistics, Inventory Control, and Supply Chain variety on supply-chain performance”, European Journal of Operational
Management”, CHOICES: The magazine of food, farm and resource Research, Vol. 143 No.3, pp. 548-69, 2002.
Issues, Vol. 20, No. 4, 4th Quarter 2005. [26]. S. Buffett and N. Scott, “An Algorithm for Procurement in Supply
[4]. R. Wilson, 15th Annual State of Logistics Report. Council of Supply Chain Management”, AAMAS-04 Workshop on Trading Agent Design
Chain Management Professionals, 2004. Available online: and Analysis, New York, 2004.
[Link] [27]. S. Behzadi, Ali A. Alesheikh and E. Poorazizi, “Developing a Genetic
[5]. R. Ballou, “Business logistics/supply chain management”, 5th Ed. Upper Algorithm to solve Shortest Path Problem on a Raster Data Model”,
Saddle River, NJ: Prentice Hall, 2004. Journal on Applied Sciences, Vol. 8, No. 18, pp. 3289-3293, 2008.
[6]. M. Fisher, “What is the right supply chain for your product?”, Harvard [28]. Aphirak Khadwilard and Pupong Pongcharoen, “Application of Genetic
Business Review, Mar/Apr., pp. 105-116, 1997. Algorithm for Trajectory Planning of Two Degrees of Freedom Robot
[7]. Guangyu Xiong and Hannu Koivisto, “Research on Fuzzy Inventory Arm With Two Dimensions”, Thammasat International Journal on
Control under Supply Chain Management Environment”, Lecture Notes Science and Technology, Vol. 12, No. 2, April- June 2007.
in Computer Science, Vol. 2658, pp. 673, 2003. [29]. M.A. Sharbafi, M. Shakiba Herfeh, Caro Lucas and A. Mohammadi
[8]. M.C. Bonney, “Trends in Inventory Management”, International Nejad, “An Innovative Fuzzy Decision Making Based Genetic
Journal on Production Economics, Vol. 35, No. 1-3, pp. 107-114, 1994. Algorithm”, World Academy of Science, Engineering and Technology,
[9]. M. Muller, “Essentials of Inventory Management”, NewYork: Amer. Vol. 19, pp. 172-175, May 2006.
Manage. Assoc., 2002. [30]. Thomas Butter, Franz Rothlauf, Jorn Grahl, Hildenbrand Jens Arndt,
[10]. S. Nahmias, “Production and Operations Analysis”, New York: “Developing Genetic Algorithms and Mixed Integer Linear Programs
McGraw-Hill/Irwin, 2004. for Finding Optimal Strategies for a Student’s Sports Activity”, Working
[11]. Krishnamurthy, Khorrami and Schoenwald, “Decentralized Inventory Papers in Information Systems, University of Mannheim, 2006.
Control for Large-Scale Reverse Supply Chains: A Computationally [31]. S.A. Qureshi, S.M. Mirza and M. Arif, “Fitness Function Evaluation for
Tractable Approach”, IEEE Transactions on Systems, Man, and Image Reconstruction using Binary Genetic Algorithm for Parallel Ray
Cybernetics, Vol. 38, No. 4, pp. 551-561, July 2008. Transmission Tomography”, International Conference on Emerging
[12]. Kanit Prasertwattana, Yoshiaki Shimizu and Navee Chiadamrong, Technologies, 2006. ICET ‘06. 13-14, Nov. 2006, pp. 196-201.
“Evolutional Optimization on Material Ordering and Inventory Control [32]. Saifuddin Md. Tareeq, Rubayat Parveen, Liton Jude Rozario and Md.
of Supply Chain through Incentive Scheme”, Journal of Advanced Al-Amin Bhuiyan, “Robust Face detection using Genetic Algorithm”,
Mechanical Design Systems, and Manufacturing, Vol. 1, No. 4, pp. 562- Journal on Information Technology, Vol. 6, No. 1, pp. 142-147, 2007.
573, 2007. [33]. M. Soryani and N. Rafat, “Application of Genetic Algorithms to Feature
[13]. Mileff, Peter, Nehez, Karoly, “A new inventory control method for Subset Selection in a Farsi OCR”, World Academy of Science,
supply chain management”, 12th International Conference on Machine Engineering and Technology, Vol. 18, pp. 113-116..
Design and Production, 2006. [34]. William M. Spears and K.A. De Jong, “On the Virtues of Uniform
[14]. “Optimization Engine for Inventory Control”, Golden Embryo Crossover”, 4th International Conference on Genetic Algorithms,
Technologies Pvt. Ltd., Maharastra, India, 2004. La Jolla, California, July 1991.
[15]. Jinmei Liu, Hui Gao, Jun Wang, “Air material inventory optimization [35]. Syswerda, Gilbert, “Uniform Crossover in Genetic Algorithms”, Proc.
model based on genetic algorithm”, Proceedings of the 3rd World 3rd International Conference on Genetic Algorithms, Morgan Kaufman
Congress on Intelligent Control and Automation, Vol. 3, pp. 1903 - Publishing, 1989.
1904, 2000. [36]. Riccardo Poli and W.B. Langdon, “On the Search Properties of
[16]. C.M. Adams, “Inventory optimization techniques, system vs. item level Different Crossover Operators in Genetic Programming”, Proceedings
inventory analysis”, 2004 Annual Symposium RAMS - Reliability and of Genetic Programming’98, Madison, Wisconsin, 1998.
Maintainability, pp. 55 - 60, 26-29, Jan, 2004.
[17]. P. Pongcharoen, A. Khadwilard and A. Klakankhai, “Multi-matrix real- ABOUT AUTHORS
coded Genetic Algorithm for minimizing total costs in logistics chain
network”, World Academy of Science, Engineering and Technology,
Mrs. S. Narmadha is working as Assistant Professor
Vol. 26, pp. 458-463, 14-16, December, 2007.
in the Department of Computer Science and
[18]. Greg Scheuffele and Anupam Kulshreshtha, “Inventory Optimization A
Engineering in Park College of Engineering and
Necessity Turning to Urgency”, SETLabs Briefings, Vol. 5, No. 3,
Technology, Coimbatore. She obtained her Bachelor’s
2007.
degree in Computer Science and Engineering from
[19]. Vecchio and Paschalidis, “Enforcing service-level constraints in supply
Tamilnadu College of Engineering, Coimbatore under
chains with assembly operations”, Proceedings of IEEE Conference on
Bharathiar University and Master’s degree in
Decision and Control, Vol. 5, pp. 5490- 5495, December 2003.
Mechatronics from Vellore Institute of Technology,
[20]. Ioannis CH. Paschalidis, Yong Liu, Christos G. Cassandras and Christos
Vellore. She is currently pursuing Ph.D. under Anna
Panayiotu, “Inventory control for supply chains with service level
University, Chennai. She has 8 years of Teaching
constraints: A synergy between Large Deviations and Perturbation
Experience and 2 years of Industrial experience. She has published 14 papers
analysis”, Annals of Operations Research, Vol. 126, pp. 231-258, 2004.
in International Conferences, 2 papers in International journals and a book on
[21]. J. Sudhir Ryan Daniel and Chandrasekharan Rajendran, “A simulation-
‘Open Source Systems’. She is life member of ISTE. Her field of interest
based genetic algorithm for inventory optimization in a serial supply
includes Supply Chain Management, Automation, Database Management
chain", International Transactions in Operational Research, Vol. 12,
Systems, Virtual Instrumentation, Soft Computing Techniques and Image
pp. 101-127, 2005.
Processing.
[22]. P. Radhakrishnan, V.M. Prasad and M.R. Gopalan, “Inventory
optimization in supply chain management using genetic algorithm”,
International Journal of Computer Science and Network Security,
Vol. 9, No. 1, January 2009, pp. 1-8.

178 [Link]
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 7, No. 1, 2010
Dr. V. Selladurai is the Professor and Head, Department
of Mechanical Engineering and Principal, Coimbatore
Institute of Technology, Coimbatore, India. He holds a
Bachelor's degree in Production Engineering, a Master's
degree in Industrial Engineering specialisation and a
Ph.D. degree in Mechanical Engineering. He has two
years of industrial experience and 22 years of teaching
experience. He has published over 90 papers in the
proceedings of the leading National and International
Conferences. He has published over 35 papers in international journals and 22
papers in national journals. His areas of interest include Operation Research,
Artificial Intelligence, Optimization Techniques, Non-Traditional
Optimization Techniques, Production Engineering, Industrial and
Manufacturing Systems, Industrial Dynamics, System Simulation,
CAD/CAM, FMS, CIM, Quality Engineering and Team Engineering.

Mr. G. Sathish is a full time research scholar under

Anna University, Coimbatore. He holds a Bachelor’s
degree and a Master’s degree in Computer Science and
Engineering. He has 8 years of industrial experience.
He has published 10 papers in the proceedings of the
leading International Conferences. His field of interest
includes Supply Chain Management, Optimization
Techniques, Data Mining, Knowledge Discovery,
Automation, Soft Computing Techniques and Image
Processing.

179 [Link]
ISSN 1947-5500