Content uploaded by Rubel Amin
Author content
All content in this area was uploaded by Rubel Amin on Sep 19, 2023
Content may be subject to copyright.
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1739
Website: www.efflatounia.com
Increasing the Efficiency and Effectiveness of Inventory Management by Optimizing
Supply Chain through Enterprise Resource Planning Technology
Rubel1, Bijay Kushwaha2
1Research Scholar, University School of Business
Chandigarh University, Mohali, India
2Associate Professor, University School of Business
Chandigarh University, Mohali, India
1rubel.r1015@cumail.in
2corresponding Author: bijay.e8413@cumail.in
Abstract
Good inventory administration is an important task in the management of the supply chain of any
business. Effective inventory handling practices aim to reduce the cost of procurement by effectively
managing and optimizing inventory, so that supply chain (SC) members are not affected by surplus and
deficit. By proper demand management, material requirement planning, sourcing, procurement, receive
and issue, we can optimize our inventory and increase the efficiency of supply chain management. To
optimize inventory, we have to control and manage running stock, excess stock, old stock, and dead
stock. ERP can help us to integrate all the processes (Demand, Material Requirement Planning, Sourcing,
Procurement, Receive, Transfer, Issue, Adjustment, ABC Analysis, FIFO Method, Safety Stock, Stock
Management) and optimize the inventory. In this paper, we propose an Inventory Optimization and Cost
Saving Model and also a method that effectively uses a Genetic Algorithm (GA) for better control of
inventory and increase the efficiency of the supply chain. This paper reports on a genetic-based approach
to improving inventory management systems in the asset management of the industry. This paper targets
specifically the quantification of the maximum level of stock and the level of deficit required for the
asset's efficiency in the transaction list so that transaction costs can be reduced. The method is used in a
three-phase delivery model that is trained to perform well. I guess that we can remove the bullwhip effect
and constraints by optimizing inventory by using ERP.
Key Words: ERP, Genetic Algorithm, Optimisation, Inventory Management, Supply Chain.
Introduction
Optimizing inventory is a technique aimed at matching capital investment limitations with service-level
areas across a big number of stock-keeping units (SKUs) through accounting for requirement and supply
unpredictability. Inventory handling software (ERP) is available from a variety of firms all over the world.
ERP may provide point solutions, such as forecasting, or certain modules that must be purchased
individually. In today's marketplaces, a company's capacity to deal with the problems of lowering lead times
and expenses, enhancing buyer service stages, and enlightening product quality is critical. By using ERP,
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1740
Website: www.efflatounia.com
we may solve all kinds of problems of inventory. Purchase, manufacture, supply, and advertising have
traditionally functioned separately. Regrettably, these organizations have different purposes, even though
they appear to be working toward the same goal. Marketing aims for a high degree of customer service and
a high volume of sales, but both goals are at odds with the goals of manufacturing and distribution.
Exploration decisions are frequently made exclusively based on lowering asset costs, while production and
distribution decisions are frequently made solely based on maximizing power while lowering production
costs (unit), regardless of greater commodity levels or longer lead times. Procurement administration entails
the successful coordination and integration of several entities with varying purposes working toward a
mutual area. Recently, the significant probable aimed at achieving these goals done good procurement
management approaches was realized. (Joines, et al, 2002).
The issue with asset management is a lack of sufficient availability of an item to fulfill the predicted
demand pattern while striking an acceptable balance between the expense of retaining inventory and the
penalty (lost sales and goodwill) at the end. This could be items sold in the store, components of machinery
left in the factory, train cars, or money in the bank to meet client needs. It is very amazing to discover that
so many seemingly disparate problems may be solved mathematically as a property management problem.
Of course, there are numerous types of innovation programs. There are three categories of associated costs.
This has a restricted value that is determined by the specific order. These include the administrative
expenses of placing an order, also known as restructuring costs or set costs; and (ii) inventory costs, also
known as inventory cost carrying costs, which include storage charges, interest, and insurance, among other
things. (iii) loss of interest, goodwill, and other costs associated with a deficit. For good procurement
management, all of the above should be prepared. (Narmadha, et al, 2009).
Holland and colleagues created the genetic algorithm in the 1960s and 1970s. The theory of evolution,
which explains the origin of species, inspired genetic algorithms. Natural selection will inevitably lead to
the extinction of endangered and endangered species in their habitats. Stronger has a high chance of passing
on genes to future generations via reproduction. Genes with the correct genetic makeup become stronger in
their genes over time. Random genetic alterations can occur while slow-moving evolution is taking place.
New species will arise from creation if these mutations provide further benefits to the struggle of existence.
Natural selection removes failed alterations.
The human or chromosomal is the GA word for the solution vector x $ X. Microorganisms are the
units that makeup chromosomes. Every element is in charge of one or more chromosomes. Genes are
supposed to be binary numbers in Holland's actual projection of GA. Various genes have been introduced
in recent investigations. In the chromosomal solution space, a particular solution resembles x in general. A
method of mapping between loci and chromosomes is required for the solution. Coding is the name for this
map. GA, on the other hand, focuses on problem-solving rather than the problem itself. GA works with
populations, which are groups of chromosomes. The populace is healthy. It all began at random. As the
search progresses, the population will develop more and more suitable solutions until it converges, at which
point it will be conquered through a sole answer. Holland also gave evidence of convergence (schema
theory) for chromosomal binary vectors' global optimism. To generate new solutions from old ones, GA
employs two techniques: crossover and mutation. GA's most essential operator is the crossover operator.
Crossover occurs when two chromosomes, commonly referred to as parents, combine to generate new
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1741
Website: www.efflatounia.com
chromosomes known as children. Parents are chosen from a population's chromosomes with a preference
for fitness so that their offspring inherit good genes compatible with their parents. It is intended that by
continually using the crossover operator, genes generated from good chromosomes will become more
evident in the population, resulting in a better overall solution. The transformer operator modifies the
signals of the chromosomes randomly. At the genetic level, modification is frequently used. The rate of
mutation (the potential for genetic mutation) during normal GA activation is relatively low and is dependent
happening the distance of the gene. As a result, the new genetic material created by mutations will be similar
to the unique. In GA, genetic alterations play a critical role. Crossover, as previously said, brings people
together by causing chromosomes in humans to match. Genetic mutations also assist the search engine to
escape to local optima by restoring genetic variation in humans. The next generation's chromosomes are
chosen during reproduction. In the most common scenario, a person's resilience determines whether the
following generation will survive. (Kalathil, 2021).
Literature Review
A network purchasing network is a complex network, with many manufacturers, many suppliers, many
vendors, and many customers. This study has created a supply chain model that considers building materials,
lead (design) lead periods, requirement and expanses information, and buyer requirements. In exchange,
the strategy creates a base-level stock in each store — a partial warehouse or end product — to lower the
initial cost of setting up the complete network and ensuring client needs. (Markus et al. 2016; Kushwaha,
2015) Companies may now proactively evaluate master and transactional data in near real-time, using the
insights obtained to patch gaps and revenue losses, thanks to rapid improvements in analytics and machine
learning (ML). In huge data sets, ML algorithms and the models are based on excel at detecting anomalies,
trends, and forecasting insights. Predictive analytics can predict demand spikes or troughs and recommend
which items, amount, and location/store should be supplied when.
A machine learning model has been created to estimate sales demand to help optimize inventory and
save money owing to high or low inventory due to erroneous demand. The model also considers return
order data to optimize returns, resulting in higher customer satisfaction and lower costs. (Archit Bansal,
2021; Kushwaha et al., 2015).
In preparing the study, the main objective was to understand the Inventory Management system at Al
Nasr Group who faced difficulties while handling materials and maintaining the workforce. While studying,
the problems which caused the delay in inventory management were identified, and also the market
potential for the gold ornaments was also valued. A comparison was also made between these 4 shops to
find their Inventory Management Efficiency and also their sales accumulation. The final result of the
analysis gives a brief idea about the company, their sales, inventory optimization, and problems faced by
them. These finding helps to suggest improvements for the Inventory management and also strategies which
will help the company to improve their profit for the future. (Talin Tenson Kalathil, 2021; Kushwaha, 2018)
A bi-level programming approach that can simultaneously decide the position of delivery centers and
the inventory plan of supply centers and retailers is used to solve the location-inventory dilemma in this
research. The findings display that cooperative optimization saves more money than individual optimization,
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1742
Website: www.efflatounia.com
that the perfect solution is more stable during demand variations, and that it meets the optimal allocation of
key consumers (Kushwaha et al., 2021b; Wanying Peng, et al, 2021).
Spare parts are kept on hand to help with product maintenance, decreasing downtime, and extending
the life of the product. The "right-to-repair" drive, which mandates that manufacturers offer enough spare
parts during the lifecycle of their items to diminish leftover and promote sustainability, has given
replacement parts inventory management a boost. Between 2010 and 2020, 148 publications on spare parts
inventory management were produced, according to this study. This review of the literature is unique in
three ways. To begin, we look at the layout of various inventory networks' supply chains to handle spare
parts. Second, depending on analytics methodologies, we divide the current study into three categories:
descriptive, predictive, and rigid analytics. (Kushwaha et al., 2021a; Shuai Zhang, et al, 2021)
The plain deliberation of inventory holding costs for the planned architecture of supply networks has
not remained extensively lectured in scholarly works. This article describes how to use a Monte Carlo
simulation model to analyze an item-by-item inventory control system to include inventory holding costs
in a supply chain optimization model. According to the findings, the SRL should only be used if uncommon
expectations are correct, and probable roles are a respectable guesstimate for computing inventory costs in
supply chain setups. (Liliana Bolaños-Zúñiga, et al, 2021)
Thanks to advances in sensors and communication technology, the company can now achieve global
data exchange and complete inventory control. Created a strategy to optimize inventory control using deep
learning. In this case, the decision-making process is based on artificial neural networks. Accepts a status
vector as input, describing the current inventory and orders in the process chain. The output is a control
vector, which manages the instructions of each station individually. In addition, for simulation-based
decision-making optimization, the incentive function is performed based on the generated storage and out-
of-stock costs. The results show that the tilt speed and scan speed are very sensitive. By significantly
reducing the total cost compared with the original state and using optimized hyperparameters to obtain
stable control behavior of the process chain of up to ten stations, the potential of the proposed method can
be demonstrated. (M. A. Dittrich, et al, 2021)
We provide a method for analyzing the optimal inventory policy while taking into account different
risk preferences of decision-makers to reduce inventory risk in supply chain systems that are subject to
uncertainty. The proposed technique uses a combination of conditional average risk and response surface
models. The proposed method uses robust optimization to solve the inventory control problem in a multi-
tier supply chain system. The simulation model is used to verify the effectiveness of the proposed strategy.
In addition, in this study, the proposed method was compared with the dual response surface method, and
the comparison results confirmed the superiority of the recommended method in improving the robustness
of the supply chain system. (Ying Yang, et al, 2021)
The study is considered significant since it is hoped that, if completed, it would provide additional
insight into stock control procedures. The study will offer an intriguing contribution to the knowledge of
the general and specific effects of store control in other private and public utilities by using a car service
facility as a reference point. In addition, the study will support the necessity to improve stock management
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1743
Website: www.efflatounia.com
and control to reap the anticipated benefits. Using the assignment technique, this research will provide a
mathematical model for optimal inventory policy to minimize overall inventory cost, and it will be applied
to a case study. (Mohamed Khalil, et, al, 2021)
Inventory is one of a company's most valuable assets, whether it's a major corporation or a small or
medium-sized enterprise. As a result, inventory management decisions have a direct impact on revenue.
The purpose of this investigation is to determine the appropriate amount of control for each inventory item
and to solve the inventory management problem of small and medium-sized businesses. In this study, we
used Ordinal Clustering (ROC) technology to aggregate inventory items from multiple items. A medium-
sized gear manufacturing company that produces 40 distinct types of planetary and custom gearboxes puts
the suggested framework to the test. According to the findings, using the recommended cluster formation
technique with ROC and quantity discounts can save you 47.64 percent on costs. This method aids in the
identification of various assemblies, the aggregation of component requirements, and the development of a
specialized inventory strategy for each component to reduce inventory carrying costs. (Ganesh B.Narkhede,
et, al, 2021)
Inventory classification is a management method used to group items with comparable characteristics
and define a set of specific control and monitoring mechanisms for each group of items. In this document,
we provide a performance-based inventory classification (PBIC) method to determine a grouping solution
for a multi-item hierarchical inventory system managed by continuous review. We believe that it is most
effective to group products based on the information contained in the value of your control strategy and the
value of performance-related parameters instead of grouping products based on the similarity of unit cost,
demand rate or delivery time. Strategy. We describe a new strategy-driven approach to establish ranking
criteria. We also use classification methods to manage the multi-dimensionality of multi-level systems to
determine one-dimensional scores. In order to improve the Pareto (ABC)-based solution, we provide a
search-based partitioning method that uses a unique aggregation method. The PBIC technique outperforms
other classification methods by a substantial margin, according to our findings. Furthermore, empirical
evidence shows that the PBIC and the optimal grouping method work nearly identically. Finally, we analyze
the impact of our findings on management, emphasizing that when managers need to conduct an effective,
comprehensive, and credible hypothetical investigation into inventory management, they must classify
problem aggregation and downscaling. (Alireza Sheikh-Zadeh, et, al, 2021)
China's big data development history is rather brief, having lasted barely ten years thus far. Even
though big data applications in real life are still limited, the supply chain has made some headway. In the
actual operation of the supply chain, various types of data will be generated. If these data can be effectively
classified and used, the "bullwhip effect" of supply chain operations will be significantly improved.
Therefore, this research proposes to use big data to build a collaborative supply chain inventory
management model and application architecture. We focus on the supply chain of the beer industry, which
is the most well-known consumer industry with a "bullwhip effect". Based on system dynamics, we built a
collaborative big data inventory management model for the supply chain of the beer industry. We use the
Vensim program for simulation and sensitivity testing. After adopting our model, we noticed that the
inventory volatility of participants in the beer industry supply chain was significantly reduced, which
proved the efficiency of the model. Our research can also be used to address potential issues with the
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1744
Website: www.efflatounia.com
collaborative inventory management approach for massive data supply chains, and it suggests some
solutions. (Jinhui Chen, et, al, 2021)
The goal of this study is to examine inventory management at PT XYZ Indonesia by optimizing the
forecasting approach. The research method employed was descriptive research, intending to use qualitative
and quantitative data to explain and describe the research object. Qualitative data defined a description of
a company's actions that were carried out through inventory management. Sales data, inventory data, and
quantity data of items arriving in the company are examples of quantitative data. The study took place in
the Storage System department over 30 months, from January 2014 to June 2016. A category was taken as
many as 1 from the ABC analysis data conducted by the company. On the article level, the least MAD value
was for the D product with linear regression forecasting method, according to the findings of research
utilizing the three types of forecasting methods. The C product has the smallest MAPE value when using
the simple moving average forecasting approach. On the level of aggregate products, exponential smoothing
forecasting produced the smallest MAD and MAPE values, implying that the exponential smoothing
forecasting approach predicts the smallest and most accurate aggregate for product deviation. (Rosalendro
Eddy Nugroho, et, al, 2021)
This study examines the contribution of the literature to the search for safety stock in the procurement
process under uncertainty and risk, focusing on the issue of scale (determining the level of safety stock).
From 1995 to 2019, we conducted a systematic literature review (SRS) in related journals, including 193
articles. The three main themes involved in these documents are safety stock size, safety stock management,
and the location, allocation, or placement of safety stock. SLR analysis reveals gaps in the literature and
research opportunities and provides a roadmap for future research on this topic. (Júlio Barros, et, al, 2021)
Ineffective inventory control of finished goods warehouses is a concern for small and medium-sized
companies in the agribusiness sector. Therefore, the importance of this project lies in improving the
inventory control of olive products. In addition, the purpose of the survey is to reduce inventory costs, since
there are currently 319,204 soles overloaded. Perfect delivery, inventory accuracy, inventory disruption,
and inventory turnover are designed to improve efficiency and standardization indicators. The FIFO method
and the PHVA method (plandoverifyact), also known as the Deming cycle, will be used in agro-industrial
organizations to standardize the processes of receiving, placing, preparing and shipping in warehouses of
finished products. (Izaguirre Malasquez, et, al, 2021).
The supply chain management strategy that underpins the policymaking progression for the purchase
of certain commodities has been taken into consideration (Buffett et al, 2004). RFQs are based on price
projections, demands, and estimates that raise the asset's value every day without lowering receivables costs.
The Markov Decision Framework (MDP) has been used to model the problem, which permits for the control
of the utilization of movements grounded on future region resources. Dynamic programming is used to
establish the acceptable pricing demands for each region in the MDP. TacT is a procurement control agent
that contains predictive, augmenting, and adaptable materials (Pardoe et al. 2009). TacTex-06's tasks
include forecasting the economy's future, such as the pricing that suppliers will offer and the grade of the
client request, and directing for forthcoming measures to guarantee extreme profitability.
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1745
Website: www.efflatounia.com
(Beamon et al, 1998). obtainable findings from the education of presentation indicators used in
procurement models, as well as a methodology for selecting the best chemical supply chain measuring
systems. Supply networks are also getting new flexibility systems. It has been hypothesized that the beam-
ACO can be achieved in asset management by (Kumar et al. 2012). Beam-ACO has been utilized to boost
supply chain shipping and logistic agents. The conventional ACO method has aided in optimizing the
dispersed scheme's performance. The adoption of Beam-ACO has improved procurement results both
locally and globally. The use of Genetic Algorithms (GA) in a profitable industry example is proposed by
(Wang et al. 2000). The GAs was used to justify the full cost of the bulk purchasing scheme in this example.
Numerous diagnostic models with stochastic needs are used to demonstrate the program. The mathematical
model was developed to illustrate a stochastic invention with many to many demands, travel constraints,
and pricing uncertainty aspects. The genetic algorithm gave its blessing by (Lo, 2007) works with the well-
known issue of production with backward realities, such as fluctuating demands over time and incomplete
production due to product interruption and clear distribution mistakes. In addition to increasing the number
of production cycles to produce an innovation policy (R, Q), an integrated production system may be created
to minimize the cost of complete collection in a given period by using production intermediate search.
[Baras et al. 2015] System Dynamics has created a simulation model for conventional retail transactions.
The goal of their simulation work was to come up with new policies that would increase income while
lowering costs for the seller. In addition, the study's goal was to look into the consequences of various
diversity techniques. A procurement model based on a stock-based system is frequently reviewed to assist
HP production managers in managing products in their supply chains (Lee et al., 2018)
(P. Radhakrishnan et. al., 2009) has created a unique and effective method based on Genetic
Algorithms to evaluate whether it is possible to minimize the cost of sales of goods by reducing surplus
stock and the amount of deficit necessary in the usage of inventory in the supply chain. The majority of
known algorithmic performance improvements have been made, however, it turns out that the majority of
them have not had the intended influence on design decisions or procurement-related issues. Some
efficiency tactics are ineffective because they are not well-suited to solving complicated performance
problems in the short time frames required for decision-making, while some problem-based strategies
necessitate advanced technology. This complicates the deployment of decision-making systems that offset
the urge towards quick adoption in a fast-changing world. IO strategies must define where acceptable assets
should be placed around the world, taking into account their cost at each stage of the supply chain, as well
as all targeted service levels and lead recovery periods.
SUPPLY CHAIN INVENTORY OPTIMIZATION ANALYSIS
Inventory Optimization and Cost Saving Model
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1746
Website: www.efflatounia.com
FIGURE 1. Inventory Optimization and Cost Saving Model
The above figure-1 represents the inventory management and cost-saving model. To make an effective
supply chain, a company or business will receive the order then make a Bill of Materials then create demand.
For made to order supply chain, a company or business will receive an order then make a Bill of Materials
then create demand. If the BoM and consumption are wrong, demand will not be accurate. For made to
stock supply chain, demand is generated by analyzing the previous history of 3-month, six months and one-
year consumption. Making accurate demand needs accurate lead time, consumption, and safety stock. If the
consumption report is wrong, demand will not be accurate. Inaccuracy of demand may happen stock out
and stock over. For stock out, we can't meet the customer requirement timely. We may lose the customer.
For stock over, we have to bear holding cost, wastage cost, and damage cost. We have to make Materials
Requirement Planning properly. At the time of MRP, we have to consider existing excess stock. Suppose
for an item we need 10000 pcs but we have 3000 pcs as existing stock then we go for purchase 10000-
3000=7000 pcs. For inventory optimization, we must utilize and maximum reuse the stock of the materials.
Then we have to go for sourcing and procurement. We have to source as per requirement quantity, quality,
and lead time. Otherwise, it will impact inventory. At the time of material receiving, we have to count and
check the quality as per requirement and documents. Otherwise, it impacts inventory by bad quality,
damage, and short quantity. Then we have to issue materials to manufacturing. Users submit issue
requisition through the system. The inventory officer checks properly requisition quantity is ok or not. Then
ERP
Accurate Demand through
system considering lead
time, accurate
consumption and safety
stock
Material Requirement
Planning Considering Excess
Stock
Receive,
counting, quality
check as per PR,
PO and ship
docs. Then entry
in ERP.
Sourcing and procurement
as per quantity, quality and
lead time of purchase
requisition
Control inventory following ABC analysis, FIFO
Method and separate running stock, excess stock
and old stock. Separate stock like as very fast
moving, fast moving, slow moving, very slow
moving and none-moving items
Completing
Inventory
optimization
and increase
the
efficiency in
supply chain
management
Check all
kinds’ reports
like demand,
PR, PO,
Receive, Issue,
transfer,
adjustment
and current
stock report
and stock
aging report
Cycle
counting: then
adjust stock in
physically and
in system
Transfer materials:
as per request
quantity and receive
transfer material as
per challan quantity.
Then transfer
through system and
receive through
system
Receive Issue Requisition from
User through System. Inventory
officer will check properly then
issue materials as per requisition
and then issue in system.
Overall Stock
Management
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1747
Website: www.efflatounia.com
they will keep ready and issue physically and in the system. We must issue materials as per consumption
and requirement of manufacturing. If we issue more than the required quantity, we have to face stock out.
If we have to manufacture in another company or subcontract, then the inventory officer will receive
transfer requisition through the system and make ready goods physically and in the system, then transfer
material to a specific location. For old stock, we have to take steps very quickly. We must reuse or sell to
others so that we can reduce our losses. By using ERP, we can integrate all of the processes. We can make
time and action plans. All the data and works are visible by ERP. We can get all kinds of the report from
ERP. Thus we can Increase the Efficiency and Effectiveness of inventory management by optimizing
Supply Chain through Enterprise Resource Planning Technology. For various reasons, inventory
optimization is critical to business operations in today's changing economic environment. The first is the
cost of carrying, maintaining, and managing inventory. Most prices are rising steadily, real estate and tax
rates are rising, and transportation costs are soaring. Inventory optimization can ensure that inventory is
balanced to meet expected demand while reducing costs and improving control over subsequent inventory
purchases. Planning reduces the need to expedite supplier orders, thereby reducing the need to expedite
customer shipments.
Second, the optimization of the inventory improves the financial performance of the inventory by
combining the purchase and storage in the demand of the expected customers. If it is not excluded, it will
help reduce the surplus stock and the accumulation of inventory in the future. 3,444 third, many known
variables that can affect your inventory are much more, and as a result, your customer service level has
your customer service level to monitor properly. I have only planned your inventory that I could not achieve
both your inventory expectation and consumer expectations. If you try to plan with ERP, it will be even
worse, as it will be worse to increase the client's misfortune and lose income.
Essentially, inventory optimization is the process of balancing an inventory's investment with a
company's fill-rate (service level) requirements. Some additional financial concerns and limits can be
applied to the algorithm creation process. These complicated algorithms operate behind the scenes so that
the user is not confused or afraid of utilizing optimization to aid accomplish their inventory.
The major goal of inventory control is to foresee where, why, and how management will be required,
and such predictions will be produced here in the process. The proposed Methodology will ensure that
adequate stock levels are maintained in the future, lowering the procurement asset's cost. The Supply Chain
model is separated into three parts, each of which will be implemented in turn.
FIGURE 2. Three Stage Supply Chain (Studied Model)
In Figure 2, The manufacturer creates a variety of products and decides how they will be delivered to
the distribution center and distributed to the agents. The Recommended Approach tries to identify the
specific product to be focused on, as well as the number of stock levels of items that different supply chain
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1748
Website: www.efflatounia.com
members must maintain, and the approach assesses whether stock quality exists. To discover the proper
value, we apply a genetic algorithm in our proposed method. Figure 3, which depicts the processes utilized
in the performance analysis, clearly depicts the performance of our method.
.
Figure 3. Genetic Algorithm Structure
Figure-3, which depicts the processes utilised in the performance analysis, clearly depicts the performance
of our method. The number of excess stock levels and the number of shares necessary for various supply
providers were initially represented by zero or non-zero numbers. Asset Control is not required if the
provider's data is zero, but it is required if the data is non-zero. Non-zero data denotes both a surplus and a
deficit. The surplus is referred to as the fair value, while the deficit is referred to as the negative value.
Human generation: Each gene consists of a series of random numbers. Here, a chromosome of three genes
with random numbers in each kind is created, as well as a product representation. Figure-3 depicts a random
human created for genetic function. The number of events in the prior record is determined after the
manufacturing of such people. This is accomplished by calculating the function () and determining the total
number of actions performed by that person in a specific product. This is the number of such instances of
the level of product shares applicable to all members for the entire period under consideration.
FIGURE 4. Chromosome Representation
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1749
Website: www.efflatounia.com
Strength performance test: In the genetic algorithm, the solution of a certain chromosome arranged in
all the other chromosomes is called an intensity function, which is a transparency function. The correct
chromosomes, or at least the closest chromosomes, can multiply and combine their data sets in one of many
ways to create a new generation that is better than we think. Fortification work issued by:
( ) ( )
log 1 , 1,2,3,..........,
occ
not
ni
f i i n
n
= − =
Where:
i= the number of occurrences of the chromosome in the record set.
n= the total number of records that have been collected from the past or the total number of data present
in the record set.
The set chromosome's enhanced function is generated at random. Following that, the chromosomes
are genetically modified.
Genetic Performance: Following the completion of the strength calculation, genetic functions are
carried out. Genetic performance can be seen through selection, crossover, and body modification.
Selection: The challenge of choosing the most powerful chromosome for the continuous genetic
function is known as the initial genetic function. This is accomplished by assigning levels to each of the
existing chromosomes based on their predicted intensity. The best chromosomes are chosen for the
continuation of the procedure based on this requirement.
Crossover: Only one crossover point is used in this study for crossover performance. The mating
lake's first two chromosomes are chosen for crossover performance. The crossover function for the model
scenario is depicted in the diagram below.
FIGURE 5. Crossover Operation
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1750
Website: www.efflatounia.com
The genes in the two chromosomes to the right of the cross-over location are exchanged, and the cross-
over procedure is completed. Two new chromosomes are obtained after the crossover procedure. The cross-
functional function is done because the correct genes for the crossing site in the two chromosomes are
altered. Two new chromosomes are discovered after crossover surgery. Genetic modification: Crossover
performance causes newly acquired chromosomes to be pushed to mutation. A new chromosome is
produced when a modification is made. This is accomplished by doing a random two-point generation and
then switching back and forth between the two genes. Figure-5 shows a diagram of the transformation
function.
FIGURE 6. Genetic Modification
The mutation process creates new chromosomes that are not identical to the ones that were created.
Following the discovery of a new chromosome, a new chromosome will be generated at random. The
previous step will be repeated using the new chromosome found in the previous step. In other words, an
excellent chromosome will be found at the end of each iteration. This will be mixed with the newly created
random chromosomes in the next iteration. Finally, among all of the persons present, the proper person is
discovered. This sophisticated chromosome provides detailed information on each supply chain member's
stock levels of a specific commodity. Based on the data, it can be inferred that a certain product and its
related inventory level play an important role in increasing the cost of a procurement project. By adjusting
the inventory level of the specific product in the appropriate members of the supply chain, sales expenses
can be reduced in the future.
RESULT AND DISCUSSION
With the help of MATLAB, the inventory management efficiency in procurement management based
on genetic algorithms is evaluated. The MATLAB script is used to generate stock levels for three different
supply chain members: factory, distribution center, and agent, and this set of data is used to assess the
genetic algorithm's performance. Table-1 shows the set of data that was used. Table 1 lists the remaining
17 data sets, which are considered historical records.
TABLE 1. A Sample of Data Sets Having Stock Levels of the Members of Supply Chain
Factory
Distribution Center
Agent
146
118
532
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1751
Website: www.efflatounia.com
-491
-239
169
372
573
-345
-491
-239
169
888
-844
208
-491
-239
169
-491
-239
169
792
-456
837
-746
721
-677
172
969
-407
-491
-239
169
611
-295
-445
-491
-239
169
482
-471
761
-992
268
-370
-152
275
-345
-491
-239
169
As mentioned above, the first two chromosomes are produced early in the genetic process. The genetic
provider's Crossover and Mutation are responsible for these first chromosomes. After cross-inserting and
adjusting our iterative value "100", the corresponding chromosome is obtained. 'After each repetition, the
chromosome obtained matches the best chromosome. Therefore, at the 100 ends, the major chromosome
"491 329 169" is found. By comparing the results obtained from the genetic algorithm with previous records,
it can be determined that the control of the chromosomes of this result is sufficient to minimize losses due
to multiple inventory reserves or inventory shortages. Therefore, it is verified that the analysis found an
inventory index, which is the best predictor of inventory in asset management.
The results of customers using advanced inventory planning and optimization show that they have
achieved (Wikipedia, 2020):
• Reduce inventory by 20% to 40% or more, usually in 6 months or less
• Express and express delivery Reduce transportation by 35% or more
• Increase productivity by reducing planning time by 60% to 80% or more
• Control costs and reduce replenishment by 15% or more
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1752
Website: www.efflatounia.com
• By 5% to 15% in customer needs Or more, provide suitable items to improve service level
More
• Out of stock reduced by 15% 30% or more
• Sales increased from 15% to 15% or more
• Due to better supplier relations, Commodity costs were reduced by 5% to 10%
We find that manually setting is a time-consuming task and quite impossible to integrate and optimize.
In our findings we can see that to optimize inventory, we have to control and manage running stock, excess
stock, old stock, and dead stock. We can also see that ERP can help us to integrate all the processes (Demand,
Material Requirement Planning, Sourcing, Procurement, Receive, Transfer, Issue, Adjustment, ABC
Analysis, FIFO Method, Safety Stock, and Stock Management) and optimize the inventory.
IMPLICATIONS OF THE STUDY
The study will provide an opportunity to gather information regarding inventory optimization. It will
also gather all the research findings and available literature in inventory optimization in Bangladesh and
South Asia. This study will focus on sustainable inventory security and financial benefits. The study will
contribute to Companies that have achieved financial benefits by employing inventory optimization.
Companies have reaped financial gains from inventory optimization, according to the research according
to a study by IDC Manufacturing Insights, several companies using inventory optimization were able to
reduce inventory levels by as much as 25% in just one year and achieve discounted cash flows of more than
50% in less than two years. By employing inventory optimization to achieve improved service levels while
lowering inventory, Electro components, the world's largest distributor of electronics and maintenance
supplies, raised revenues by £36 million. In two years, Castrol used inventory optimization to cut finished
goods inventory by an average of 35% while enhancing service levels (measured as line fill rates) by 9%.
Smith's Medical, a Smiths Group company, employed inventory optimization to better meet demand
volatility and supply unpredictability, decreasing the risk of understocking and overstocking while
smoothing out manufacturing cycles. (Wikipedia, 2020) In every business and company inventory
optimization by ERP is applicable and they will be benefited from cost-saving and time savings.
LIMITATIONS
The scope of inventory optimization is very broad and includes many theories on how to
evaluate the chain. This article will not go into detail about everything that is included in the term inventory
management. The purpose of this article is to describe the methods that can be used to determine whether
inventory optimization is effective and to provide a model or index that combines multiple data. Inventory
optimization practices vary based on company size, company business, and rules implemented by local
governments, and vary from company to company. The second part of the empirical study will be carried
out in a single company or a few, and cannot be considered representative of all companies. This means
that the results of this research must be taken into account. Due to the need to protect confidential
information, the results of this phase of the empirical investigation must be presented without real numbers.
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1753
Website: www.efflatounia.com
CONCLUSION
The main reason for choosing the 'Inventory Optimization' as the title is that Inventory Management
is a very important factor to every company and a company with inefficient Inventory Management may
lead to difficult management and loss of the customers. The findings show that collaborative optimization
saves more money than individual optimization, that the perfect solution is more stable during demand
variations, and that it meets the optimal allocation of key consumers. On the other hand, inventory costs
can have a significant impact on the expansion or contraction of the optimal structure of the supply chain
and distribution system. The latent function is a good approximation for calculating inventory costs in a
supply chain environment.
The subject of inventory management is fundamental to the performance of any organization and is
one of the most important factors for its long-term viability and high productivity. The most effective
inventory management is to reduce costs as much as possible while meeting customer expectations for
services such as delivery accuracy and lead time.Inventory management is an important part of supply chain
management. We discussed cost-saving and inventory optimization models and gene-based algorithms to
maximize asset allocation in supply chain management, and focused on how to directly determine the
amount of inventory required and the total cost when reducing inventory. Operations in the supply chain.
We apply our method to the inventory optimization model and the three-step learning model for
improvement. The proposed method was implemented and MATLAB was used to test its effectiveness.
Manually setting reorder settings (such as safety stock or minimum/maximum levels) is a time-consuming
task. Therefore, it is rarely completed and cannot adapt to changing demand conditions. These activities
should not consume the valuable time of the buyer/planner. The re-ordering (replenishment) settings can
be optimized to save a lot of money. Because ERP can accomplish it quickly, the results dynamically reflect
changes in demand.
REFERENCES
1. Bask, A., Lipponen, M., Rajahonka, M., & Tinnilä, M., The concept of modularity: diffusion
from manufacturing to service production. Journal of Manufacturing Technology Management
Vol. 21 No. 3, 2010 pp. 355-375 (2010).
2. Barros, J., Cortez, P., & Carvalho, M. S., A systematic literature review about dimensioning safety
stock under uncertainties and risks in the procurement process. Operations Research Perspectives,
100192 (2021).
3. Bolaños-Zúñiga, L., & Vidal-Holguín, C. J., The impact of inventory holding costs on the strategic
design of supply chains. Revista Facultad de Ingenieria Universidad de Antioquia, (101), 45-54
(2021).
4. Buffett, S., & Scott, N., An algorithm for procurement in supply chain management. In AAMAS
2004 Workshop on Trading Agent Design and Analysis (2004, July).
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1754
Website: www.efflatounia.com
5. Beamon, B. M., Supply chain design and analysis:: Models and methods. International journal
of production economics, 55(3), 281-294 (1998).
6. Bansal, A., Role of Machine Learning in Inventory Optimization using Time-series Forecasting
(No. 5930). EasyChair, (2021).
7. Chen, J., & Jin, C. Y., A Study on the Collaborative Inventory Management of Big Data Supply
Chain: Case of China's Beer Industry. Journal of the Korea Society of Computer and
Information, 26(3), 77-88 (2021).
8. Dittrich, M. A., & Fohlmeister, S., A deep q-learning-based optimization of the inventory control
in a linear process chain. Production Engineering, 15(1), 35-43 (2021).
9. Ettl, M., Feigin, G.E., Lin, G.Y., and D.D. Yao, A supply network model with base-stock control
and service requirements. Operations Research, Vol. 48, pp. 216–232 (2000).
10. Joines, J. A., Gupta, D., Gokce, M. A., King, R. E., & Kay, M. G., Supply chain multi-objective
simulation optimization. In Proceedings of the winter Simulation Conference (Vol. 2, pp. 1306-
1314). IEEE (2002, December)..
11. Khalil, M., Ellaimony, E. S., Taha, A., & Osman, M. Proposed Model for Optimal Inventory
Policy, International Journal Of Scientific Advances, 2 (1), 21-29, 2021.
12. Kalathil, T. T., Inventory optimization for a company with luxury and high valued products and
services (2021).
13. Kumar, T., Singh, S. R., & Gupta, C. B., Genetic algorithm based multi-product and multi agent
inventory optimization in supply chain management. International Journal of Modeling and
Optimization, 2(6), 653 (2012).
14. Kushwaha, B. P. (2015). An empirical study on youth perceptions and preferences on smart phones.
International Journal of Research in IT & Management, 5(10), 82-90.
15. Kushwaha, Bijay Prasad; Rao, N. Sambasiva; Ahmad, Sohail Yasin (2015). The factors
influencing consumer buying decision of electronic products. Management Dynamics. 2015, Vol.
15 Issue 1, p5-15. 11p.
16. Kushwaha, B. P. (2018). Sustainable Offering Practices through Stakeholders Engagement.
Journal of Innovation for Inclusive Development, Vol. 3, NO. 1, 22
17. Kushwaha, B.P., Tyagi, T., Shiva, A. (2021a). Investigating the role of reinforcement and
environmental factors in balancing the state of apprehension: evidence from India. World Review
of Entrepreneurship, Management and Sustainable Development, 2021 Vol.17 No.2/3, pp.142 -
160.
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1755
Website: www.efflatounia.com
18. Kushwaha, B.P., Singh, R.K., and Tyagi, V. (2021b). Investigating Privacy Paradox: Consumer
Data Privacy Behavioural Intention and Disclosure Behaviour. Academy of Marketing Studies
Journal, Vol. 25, No. 1, pp. 1528-2678-25-1-353.
19. Levi, R., Martin Pal, R. O. Roundy, and Shmoys, D. B., Approximation algorithms for stochastic
inventory control models, Mathematics of Operations Research, Vol. 32, pp: 284-302 (2007).
20. Lee, S. H., Dekker, I. R., & Colak, A. E. Memetic algorithm for stochastic inventory optimization
with seasonal demand (Doctoral dissertation, Master thesis. Erasmus School of Economics,
Erasmus University Rotterdam) (2018).
21. Liu, J., Gao, H., & Wang, J., Air material inventory optimization model based on genetic algorithm.
In Proceedings of the 3rd World Congress on Intelligent Control and Automation (Cat. No.
00EX393) (Vol. 3, pp. 1903-1904). IEEE (2000, June).
22. Lo, B. S. Inventory optimization in an aerospace supply chain (Doctoral dissertation,
Massachusetts Institute of Technology) (2007).
23. Matei, I., Gueye, A., & Baras, J. S., Flow control in time-varying, random supply chains.
Transportation Research Part E: Logistics and Transportation Review, 77, 311-330 (2015).
24. Narmadha, S., and Selladurai, V. Multi-factory, Multi-Product Inventory Optimization using
Genetic Algorithm for Efficient Supply Chain Management, International Journal of Computer
Science and Network Security, Vol. 9, No.12, pp. 203-212 (2009).
25. Nugroho, R. E., & Resodiharjo, M., Inventory Management Analysis by Optimizing the
Forecasting Methods (Case Study At Pt Xyz Indonesia). Dinasti International Journal of
Management Science, 2(3), 435-455 (2021).
26. Narkhede, G. B., & Rajhans, N. Optimizing Inventory Carrying Cost Using Rank Order Clustering
Approach for Small and Medium Enterprises (SMES). Journal of University of Shanghai for
Science and Technology, Volume 23(1), pp. 161-170 (2021).
27. Peng, W., & Qin, L., Collaborative Optimization for Location-Inventory Problem of Commercial
Vehicle Distribution Network. In IOP Conference Series: Earth and Environmental Science, Vol.
638, No. 1, p. 012006 (2021). IOP Publishing.
28. Podobnik, V; Ana, P. and Gordan, J., An Agent-based Solution for Dynamic Supply Chain
Management. Journal of Universal Computer Scienc, Vol. 14, No. 7, pp. 1080-1104 9(2008).
29. Radhakrishnan, P., Prasad, V. M., & Gopalan, M. R., Inventory optimization in supply chain
management using genetic algorithm. International Journal of Computer Science and Network
Security, 9(1), pp. 33-40 (2009).
Efflatounia
ISSN: 1110-8703
Pages: 1739 – 1756
Volume: 5 Issue 2
1756
Website: www.efflatounia.com
30. Singh, S. R., & Kumar, T., Inventory optimization in efficient supply chain
management. International Journal of Computer Applications in Engineering Sciences, 1(4),
(2011).
31. Sheikh-Zadeh, A., Rossetti, M. D., & Scott, M. A., Performance-based inventory classification
methods for large-Scale multi-echelon replenishment systems. Omega, 101, 102276 (2021).
32. Yang, Y., Tang, L., & Ma, Y., Robust parameter design of inventory policy considering the risk
preference of decision makers. In IOP Conference Series: Materials Science and Engineering, Vol.
1043, No. 4, p. 042029 (2021). IOP Publishing.
33. Zhang, S., Huang, K., & Yuan, Y., Spare Parts Inventory Management: A Literature Review.
Sustainability, 13(5), 2460 (2021).
