See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/352235223

Inventory management for retail companies: A literature review and current

trends

Conference Paper · March 2021

DOI: 10.1109/ICI2ST51859.2021.00018

CITATIONS READS

0 7,583

4 authors, including:

Rodrigo Arcentales-Carrion Mario Pena

University of Cuenca University of Cuenca
13 PUBLICATIONS   21 CITATIONS    32 PUBLICATIONS   77 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Análisis y definición de estrategias para el desarrollo de sistemas de mantenimiento industrial View project

Feature engineering based on ANOVA and cluster validity assessment for fault diagnosis in bearings. View project

All content following this page was uploaded by Mario Pena on 13 September 2021.

The user has requested enhancement of the downloaded file.

 Inventory management for retail companies: A
literature review and current trends
Cinthya Vanessa Muñoz Macas Jorge Andrés Espinoza Aguirre Rodrigo Arcentales-Carrión
Industrial Engineering, Faculty of Industrial Engineering, Faculty of Research Group in Accounting,
Chemical Sciences Chemical Sciences Finance, and Taxation, Faculty of
University of Cuenca University of Cuenca Economics and Administrative Sciences
Cuenca, Ecuador Cuenca, Ecuador University of Cuenca
0000-0001-9820-0331 0000-0001-5685-2681 Cuenca, Ecuador
0000-0002-9700-8898

Mario Peña
Research Department (DIUC)
University of Cuenca
Cuenca, Ecuador
0000−0002−3986−7707

Abstract—In recent years, the correct management of imply being able to minimize the risk of contracting results
inventories has become a fundamental pillar for achieving that may put the health of the company at risk.
success in enterprises. Unfortunately, studies suggesting the
investment and adoption of advanced inventory management Various technologies have been developed over time for
and control systems are not easy to find. In this context, this inventory management, going from basic manual reporting to
article aims to analyze and present an extensive literature an integrated information system (IS), which can help to
concerning inventory management, containing multiple “decide how and where orders should be fulfilled to improve
definitions and fundamental concepts for the retail sector. A service levels while decreasing total costs” [6]. Moreover,
systematic literature review was carried out to determine the these new functionalities can collaborate in the most effective
main trends and indicators of inventory management in Small handling of materials and better manage the cycle of purchase
and Medium-sized Enterprises (SMEs). This research covers - reception - allocation in production [7], [8].
five years, between 2015 and 2019, focusing specifically on the
retail sector. The primary outcomes of this study are the leading This article aims to present an extensive literature review
inventory management systems and models, the Key concerning inventory control and management in the retail
Performance Indicators (KPIs) for their correct management, sector. First, the paper includes a systematic literature review
and the benefits and challenges for choos ing or adopting an regarding the Key Performance Indicators (KPIs) of inventory
efficient inventory control and management system. Findings control and management in retail companies. Second, the
indicate that SMEs do not invest resources in sophisticated main systems, methodologies, and tools used for inventory
systems; instead, a simple Enterprise Resource Planning (ERP) management are described. Finally, the current trends in
system or even programs such as Excel or manual inventories inventory handling and management in retail companies are
are mainly used. outlined. For this, the application of the Fink [9] and the
Population, Interventions, Controls, and Outcome (PICO)
[10] methodologies were developed, which suggests different
Keywords—retail, management, inventories, smes, literature steps and stages, to solve the problems and research questions
review raised. To answer these questions, a great variety of digital
I. INTRODUCTION databases were used. Thus, conference and journal articles
concerning inventory control and management in retail
Nowadays, organizations, and especially those performing companies were retrieved. The articles selected were analyzed
activities in the retail sector, face multiple challenges in the through the Atlas.ti software. Finally, a qualitative and
planning and management of their resources. For this sector, quantitative analysis was performed to answer the research
having efficient management of human, technological, or questions raised. The organization of the paper is as follows.
material resources refers to the performance that companies Section 2 presents the methodology through which the work
characterized by the experience gained in their management was guided. The next section encompasses results together
could obtain over time. Therefore, the correct inventory with analysis and discussion. Ultimately, Section 4 shows the
management has become essential, especially in organizations conclusion section gathering the main findings as well as
dedicated to retail [1-3]. The determination of the optimal proposals for future work.
inventory level is a fundamental part of the life of
organizations due to the high investment that it represents at II. RESEARCH METHODOLOGY
the time of its acquisition, administration, and maintenance. To accomplish the goals described above, the study
According to [4], [5], “the role of inventory management is to follows the Fink methodology [11]. It consists of seven main
ensure that stocks of raw material or other supplies, i.e., work- tasks: (1) choosing research questions, (2) defining
in-progress and finished goods, are kept at levels that provide bibliographic or article databases, (3) selecting search terms,
maximum service levels at minimum costs”. This because the (4) applying practical screening criteria, (5) applying
realizable asset occupies a significant percentage within the methodological screening criteria, (6) doing the review, and
Total Assets. Hence, its correct ordering and administration (7) synthesizing the results. Therefore, the systematic

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

literature review starts establishing the particular needs for metadata analysis are presented. Finally, the research
knowledge or research questions. For this purpose, the PICO questions are answered in the second part of this Section.
methodology was used [12]. In this case, Population refers to
retail companies; Intervention relates to inventory control and A. Metadata Analysis
management; Comparison refers to identifying systems, Based on the literature review, several results were
methodologies, and tools, and Outcomes refers to answer the obtained with the codes used. Within these codes are
research questions. Therefore, the research questions demographic data, such as year and country. In the results
addressed in this study are the following: What are the main obtained between the years 2015 and 2019, it can be seen that
KPIs of inventory control and management in retail studies have been increasingly carried out in such a way that
companies? What are the systems, methodologies, and tools in Fig 1, it is observed that 34.21% of the investigations were
aimed at inventory control and management in retail published during the year 2018 and 15.79% during the years
companies? and What are the current trends in inventory 2015, 2016, and 2019. It is worth mentioning that during 2019
control and management in retail companies? there is a reduction in publications since not all publications
are covered due to the period in which this study is carried out.
As a second step, a relevant set of digital databases were Also, in Fig 2, it is observed that 23.68% of studies were
selected. It was decided to use both specific and general digital performed in the United States, being the year 2018 in which
databases, namely: Emerald, Science Direct, Scopus, and the majority of publications were made in this country,
Taylor & Francis. Based on the research questions, the search 18.42% in the United Kingdom, and 15.79% in the
terms were defined (Step 3) and the searching string was Netherlands. Finally, countries with less number of reported
established as follows: retail AND inventory AND SMEs studies are China, Hong Kong, India, Poland, Singapore,
AND (model OR management OR industry OR technology Switzerland, Taiwan, Vienna, and France, all with a
OR cost OR control OR maintenance OR system OR contribution of 2.63%.
optimization OR rotation OR turnover OR software OR trend
OR tendency OR tool OR tics. This chain searching was
examined in the title or abstract document, and the selection
was limited to conference and journal articles, in English and
the period 2015-2019 (Step 4). It is worth noting that duplicate
articles were removed. A numerical summary of the potential
articles is presented in Table 1.

TABLE I. SEARCH RESULT S IN BIBLIOGRAPHIC DAT ABASES

Database Number of articles Relative percentage

Emerald Insight 39 13%
Science Direct 167 54%
Scopus 99 32% Fig. 1. Analysis of Articles by Year.

Taylor and Francis 7 2%

Total 312 100%

The final sample articles consisted of 312 potentially

relevant studies. The full text of each document was retrieved
for detailed evaluation to discard those that did not fulfill the
selection criteria with the inventory management and control
topic in retail companies. To determine the article's
contribution, the main sections, such as abstract, results,
discussions, and conclusions, were analyzed in-depth (Step 5).
As a result, 42 primary studies were subsequently selected for
doing the review using the Atlas.ti software (Step 6). In each
article, information extracted was the following: (1) Fig. 2. Analysis of Articles by Year and Country/City
demographics information (title, authors, journal name,
country and year), and (2) information related to the study B. Content Analysis
(enterprise activity, implementation sector, theoretical
foundation). The data were collected thought suitable codes In the succeeding part, the qualitative results of the study
such as technology, inventory type, cost, stock type, order based on the research questions posed are presented.
form, control method, reposition, information levels, 1) What are the main indicators (KPIs) of inventory
planning, and policies. This information was used to construct control and management in Retail companies?
code nets and to establish the main topics of this contribution. Based on the literature review regarding indicators of
Finally, a quantitative (metadata analysis) and qualitative inventory control and management within retail companies,
(content) analysis was performed (Step 7). 22 key aspects have been identified to consider when
III. RESULTS AND DISCUSSION performing inventory management, which could play the role
of performance indicators or performance within inventory
In this section, the results of the systematic literature control and management decisions. All the KPIs identified
review and current trends in inventory management for retail allow knowing the effectiveness of inventory control and
companies are reported and discussed. First, the results of the management carried out within retail companies. Among
them, the first ten are considered essential due to its higher control, the inaccuracy of inventory records on various
frequency of employment, while the remaining 12 have a products, and the lack of inventory review [23] [15] [24].
minimum frequency, as shown in Fig. 3. The KPIs identified Finally, low inventory levels at some point can be leveraged
are: 1) actual inventory and its relationship with the company's to carry out marketing strategies focused on scarcity [22].
information system, 2) inventory level, 3) shortage of scarcity,
4) product reordering, 5) product replenishment, 6) service d) Product reordering or replenishment.
level, 7) products availability, 8) excessive inventory, 9) items In the retail environment, preventing the loss of sales
on the shelf, 10) income level – earnings, 11) preference and opportunities requires the accurate and timely replacement of
purchase decision, 12) response level, 13) lost items, 14) products to customers [25]. A periodic replenishment policy
returns, 15) complexity and operational performance, 16) is essential, which is based in the variation of the levels of
proper planning, 17) inventory balance, 18) sales data, 19) services and costs under the specific policies established [26].
verification of the amounts received in stores, 20) incorrect Failing to perform a timely replenishment, the products are
scanning at the time of payment, 21) incorrect deliveries, and, considered depleted. It is possible to have inventory in the
22) adaptive, flexible inventory. backroom and still have an empty shelf, thus causing a loss of
sale [27]. Finally, it is essential to emphasize that the total cost
a) Actual inventory and its relationship with the of a product may increase due to short delays in its
company's information system. replacement [23].
A retail store must have the same data in all its records,
that is, the data that has been recorded in the information e) Service level.
system must be the same data that is physically held [13]. This Poor service level results from having inadequate
is considered necessary due to continuous inconsistencies that inventory control parameters [28]. A higher level of service is
exist between the physical inventory record and the inventory obtained if there is a decrease in the level of unsatisfied
that appears in the system, incurring operational consequences demand and if the levels of the finished product at the
[14]. Furthermore, this inaccuracy of inventory registration beginning of a period increase [19]. The level of inventory
causes an increase in the number of shortages [15] and service for seasonal products should be kept high for basic
damages the operating performance and the reputation of the clothing fashion retail companies. Conversely, the fashion
retailer with consumers [16], which could be improved by manufacturer can reap benefits by improving its profit level if
providing inventory information to the customer. In this way, the inventory service level target is low. Also, it is best to
the level of quality of service received by the customer can be maintain a low inventory service level on certain products in
projected and, in turn, the existence of the product is ensured the luxury fashion market as it projects product exclusivity
[17]. Another benefit of maintaining an inventory information [29].
system is to know how much physical inventory the company f) Availability of products.
has and how much stock was diverted from the records [15].
The product availability is related to the inventory
Ultimately, a retail store has monetary losses as a direct result
information provided to the customer, through which the
of the difference between the information recorded on actual
customer verifies the service quality [17]. Besides, this
inventory and the record in the company's information system
[13]. information influences the customer's decision when the
purchase is made [22]. Offer sellers assign a fixed amount of
b) Inventory level. inventory to a sale, and the offers page shows in real-time the
When an assortment planning problem arises, it is percentage of products claimed. This reveals the availability
essential to develop an optimization model that considers the of products to the customer. By doing so, the inventory level
optimal assortment and the inventory level [18]. At the consumed would generate a desired effect of the product, and
beginning of a season, the inventory level must be lower to thus more sales of popular products would be made. In this
avoid that in the end it is sold at a low price, this is a way, inventory and sales control are maintained [17].
consequence of keeping the product at a fixed level throughout g) Product replacement.
the season [19]. Therefore, maintaining a low inventory level
Among the strategies used by retailers to minimize the
of a product can create pressure among customers causing the
effect on operational activities caused by inventory, several
immediate purchase of the product [17]. Conversely, as the
different errors can be detected, such as storing additional
number of inventory units increases, the difficulty of detecting
items or increasing the frequency of restocking of stores with
inventory discrepancies also increases [16]. One reason to
maintain an adequate level of inventory is that, in case of the purpose of maintaining a high level of inventory [15].
More frequent product replenishment results in reduced space
requiring to make a transfer, this allows the decision on the
needed for each item, warehouse release, and increased shelf
ideal quantity [20]. Finally, it would be interesting to consider
space; consequently, it ends up managing a greater variety of
the effect that the inventory level has on the profit levels [21].
products and freeing up capital [26]. Furthermore, a decrease
c) Shortage of scarcity. in the replacement rates of products affects the optimal level
A lack of products can be caused by various factors, of inventory, causing its increase [30].
including differences between product costs, which creates the h) Excessive inventory.
possibility of a shortage of an expensive product and an excess
of cheap products [13]. There is an interaction effect between One of the strategies retailers use to deal with excess stock
is to apply product rebates or pricing. The discounts applied
scarcity levels and price leadership; when it is the case that the
to the products will depend on the deterioration rate and the
scarcity increases exceeding a certain limit, the effect is
useful life of the product. The possibility of excess inventory
reestablished. When a high level of shortage at the time of sale
increases if a timely inventory level review is not carried out
occurs, there is also an increase in the probability of the sale
order being returned [22]. One of the reasons for a shortage is [24]. Consequently, excess inventory is directly caused by
inappropriate inventory control [28].
the inefficiency in the inventory replacement process and
 i) Items on the shelf, Response level, and Income level 2) What are the systems, methodologies, and tools aimed
– earnings. at managing and optimizing inventory in Retail?
The importance of a response, relative to inventory, is Through the literature review, the main systems, methods,
greatest when it comes to reducing the risk of low profit levels and tools used for inventory management were identified.
for the fashion retailer and its supply chain (SC). When the This study considered them as systems allowing better
inventory service is higher, it is required that the quick inventory control, management, and supply. Based on their
response must be also higher [29]. The presence of an item on focus areas, three groups were created: inventory records and
the shelf allows a quick response and a higher probability of localization, optimization, and the order quantity
selling the item [13]. Hence, this aspect is essential in a retail determination.
store since higher turnover is assured in the inventory level,
a) Systems, methodologies, and tools focused on
having an impact on the level of profit or income [21] [31].
inventory records and localization.
j) Lost items, incorrect scanning, and verification of In this group, bar code and Radio Frequency Identification
the amounts received, and complexity and operational (RFID) were identified. Unlike price labels where the
performance. checkout employees have a chance to validate the cost with
Retailers suffer from product misplacement problems; the the product, bar codes are not commonly used to control the
reason could be that the product is sent to wrong storage product-specification created out of the bar code scan with the
shelves [32]. Having lost articles gives the possibility of lost existent product. To evade ticket switching, it is required the
sales because the products are misplaced [25]. The integration of the product weight next to its cost and additional
concurrence of lost items is one of the causes of inventory data in the payment method. In addition, the staff is needed at
inaccuracy [33]. Other reasons that affect the registration of the checkout counter to scan the product's barcode and
inventories are the incorrect scanning of products at the time position the product on the machine in order to measure its
of payment and verification problems of quantities received in weight. If the product weight is considered into the outlet
the store that frequently occur due to operational failures [15]. protocol, an alert can be triggered by any relevant deviation in
Finally, inaccuracy in the inventory record affects the existent and measured weight [13].
operational performance of a retailer [16]. Price calculations can be presented in real-time.
k) Preference, Purchase Decision - Sales Data, and Additionally, it offers the store the opportunity to know the
Inventory Balance. actual stock count information within fixed readers. RFID
Customer preferences, opinions, and purchasing decisions stock counts (by stationary or handheld readers) allow
can be learned through inventory information [17]. inventory levels to be evaluated every day considering each
Companies can take advantage of this information because the stock line in every area of the store. This item-level tagging
sales data reflects the customer's feelings and preferences, and tool is able to reduce the technology breach and give the
based on these data, inventory management can be improved retailers both the accuracy and the ease of use which are
[34]. The performance of an inventory system can be needed in order to help their merchandising plan and store
determined when calculating the probability of the inventory display performance. This will provide strategic information
equilibrium state [30]. which can be used for different decisions, for example ‘money
mapping’ of store sections, which highlight ‘hot zones’ where
l) Incorrect deliveries and Returns. the sales of the displayed products are quicker and in a greater
One of the causes of inventory inaccuracy is making amount in comparison to ‘cold’ zones, which are
incorrect deliveries, driving an increase in the return of characterized by slower stock lines [36]. In the retail store,
products [33]. This creates pressure on the operating resources there are four main ideas on which the procedures are based
managed by retailers. Another critical aspect is the level of on: improving stock exactness, out of stock management,
shortage, if this is high it will cause an increase in the products localization, and loss recognition. The benefits of
probability that the customer will return the order [22]. this tool include; field of vision is not needed for the item
registration, various products are able to be registered with a
m)Adaptive, flexible inventory, and proper planning. single can, also tickets can be read from quite a great range.
Inventory problems imply low adaptability and a lack of The location of the system is important due to its influence on
functionality in the retailer’s SC [35]. These SC processes the coverage it provides, while the front system offers a better
require adequate preparation due to their complexity [1]. coverage facing the products, the upper system improves
Determining adequate amount of raw materials and finished coverage at the top of the shelves. The top system is found to
products contributes to minimizing inventory costs [19]. be more effective in comparison with the front system (where
the scanner is in front of the item) as a result of the diverse and
random label orientation [37] . It is important to define the
intended purposes of meeting the application requirements
since the most extended coverage scope given by this
technology is not required. Therefore, it is achievable to evade
its major disadvantages concerning costs, tasks maintenance,
and applying proper protocols [38]. RFID enables the
restructure of inner procedures in the store to hold the chance
to improve the operations efficiency considering service
levels and total costs, relying on ticket costs and information
quality [23]. RFID plays a strategic role under competition by
focusing on the impact caused by unit tagging cost in the
manufacturer-retailer interaction through the processes of
Fig. 3. Frequency of Mention of Indicators in Inventory Management.
hiring and wholesaling pricing [32]. Another technology that constant synchronization and improved information accuracy
includes an antenna is RAIN RFID, which decreases costs at [2].
the same time that upgrades the range of vision, product
Ultimately, the Threshold Accepting algorithm has been
tracing, exactitude, identify losses, and allowing to find
missing items [37]. Besides, the Tree-based protocol used to resolve a variety of optimization problems due to its
organizes the IDs by using the length of a ticket ID as the robust optimization heuristic. The costs for the manufacturer
length of the tree, then it uses ‘0’ and ‘1’ in for the left and implicate production costs, whereas the costs for the retailer
right side of the tree respectively. A scanner searches for ‘0’ implicate shortage, stock holding, fixed ordering, and buying
costs [42]. It is important for firms to define their decision
and all the tickets whose IDs start with ‘0’ reply. If it’s a
favorable run (i.e., one ticket replies) or an empty run (i.e., no variables aiming to increase their profit. The manufacturers
ticket replies), the scanner searches for ‘1’ and all the tickets should define the prices related to large scale sales, and the
whose IDs start with ‘1’reply, this procedure goes on until all retailers should be able to define their replenishment cycles
the tickets have been recognized giving as a result a binary and their prices as well. Differential Evolution algorithms are
also presented in order to deal with the second level of the bi-
tree. The Iterative ID-free Protocol (IIP) recognizes the
presence of an item as a result of a reply given by a hash level matter, as an alternative to solve the second level of the
function, which pseudo-randomly chooses a space in a time bi-level problem, where both manufacturers and retailers
period searching for the ticket ID. Missing tickets can be should determine their correspondent price and also in the
retailers case, determine the replenishment cycle [42].
detected as a result of the comparison between the expected
and the observed data. Therefore, if the reader confirms that c) Systems, methodologies, and tools focused on
exists an empty space where it supposed to be a ticket, the determining the order quantity.
scanner determines that the ticket is missing. The Slot Filter- In this group, the main elements found are the following:
based Missing Tag Identification (SFMTI) resolve the Integrated Information System (IIS), Economic Order
predicted collision spaces into singleton spaces and separates Quantity (EOQ), Joint Replenishment Problem (JRP), Vendor
the predicted empty spaces along with the unsolvable collision Managed Inventory (VMI), the Ordinary Empirical (OE)
spaces, improving as a result the expected time efficiency. Distribution, Inventory Record Inaccuracy (IRI), Markov
This protocol shows an efficient collision solving method, decision process (MDP), All Unit Discount (AUD) policy and
which can take collision spaces and turn them into the Incremental Quantity Discount (IQD) policy, and Fuzzy
predicted singleton spaces, in addition the predicted empty Inventory Management.
spaces and the unsolvable collision spaces are separated and
not carried out for the purpose of achieving time efficiency IIS improves services levels and reduce costs by choosing
[39]. An application of RFID is called Smart Shelf; it can how orders have to be fulfilled as well as where they need to
register and detect the position of the products placed, be fulfilled. Additionally, it should be able to increase the
previously labeled with RFID tags [40]. inventory visibility across multiple nodes [43]. It also allows
reserving the stock, prioritizing and tracking client’s orders,
b) Systems, methodologies, and tools focused on and managing return flows. IIS could improve material
optimization. handling effectiveness and efficiency in each node [2]. The
The main elements identified in this group were the global challenge for IS focus on the support of information
Bayesian estimation method, the Multi-Channel Distribution sharing as well as fast decision making involving several
Center (MCDC), the Logistic Information Systems (LIS) for actors [2].
Omni-channel, and the Threshold Accepting Algorithm. The
Bayesian method estimates the error distribution caused by EOQ is an order policy focused on deteriorating products.
inspections using finite information out of periodic shelf It determines the best restocking time as well as the period of
inspections. Its goal focuses on providing strategic parameters time in which the stock will drain completely. The best
to managers to make empirically informed decisions. backlogging and ordering policies can be determined using
Management scientists have used this method in order to these values in order to minimize the total stock cost per unit
deduce demand metrics and stock levels to improve time [44]. A partial delay in the payment is permitted only if
replenishment decisions [41]. the store’s order size is bigger than a minimum amount. If the
minimum quantity is considerable, the store is going to have
MCDC is defined as an inventory system that involves preference for a partial payment delay. The advantage for the
cooperation between workers, Warehouse Management store consists in having a courtesy period to complete the
System (WMS) hardware and software, items and their layout, payments after an initial partial payment for receiving the
and stock policies implemented. This optimization model items. Applying this method, the store total cost is higher
needs accurate, available information about daily inventory considering the ordering cost; more storage cost rises the total
involving both items and locations to improve them for cost at the same time that reduces the backorder level and a
costumers and retail orders [16]. lower total cost is achievable due to an stretched delay period
The continuously evolving omnichannel panorama affects [44]. An EOQ model gives the retailer the opportunity to
the back-end LIS which helps material handling related to analyze the union between order size and expected costs [45].
warehouse activities ensuring an efficient and effective JRP defines if several items are ordered together or not. In
process, in addition it also helps with order and inventory order to reduce the inventory system total cost, it is necessary
management in the network. The progress of LIS with a proper to consider two elements from the ordering problem
functionality is associated with a well-developed Omni- considering an inventory with various items: major ordering
channel logistics. Essential aspects to contemplate to make (fixed, setup) and minor ordering costs. The former arises
Omni-channel logistics are adaptable platforms which support when the order is made no without considering the number of
adding nodes and updating software, increased integration, diverse items in the order. The latter is used considering every
singular product in the reposition order [14].
 VMI is a SC application where the responsibility of the models. In the study carried out, 22 relevant indicators are
inventory management at the retail point, belongs to the established, 10 of them show a higher mention frequency
supplier. A supplier is in charge of choosing the right time and within various studies analyzed. The importance of these
quantity to order on behalf of the retailer [46]. Implementing indicators is given as a result of the analysis presented above,
VMI based models allows Retailer Centers (RCs) to avoid in which each indicator is displayed and shows its influence
stock decisions and hold stocks, therefore letting the supplier over adequate inventory management. Furthermore, these
to define on both decisions, stock and routes. In this way, a indicators will allow the organization to know the state of the
general SC optimization is achieved, benefiting both the RCs inventory, as well as the model evaluation, which can be used
and the supplier [47]. Implementing VMI with a lack of to make continuous improvements to contribute to the
appropriate replenishment policies can make this alliance fail business wellness.
[46].
On the other hand, the current tendencies in inventory
The OE Distribution method is applied in order to model management are primarily focused on developing tools that
slow-moving demand operations, as well as working with enable retailers, product location control, loss detection, stock
time series and parameter constellations due to its great management, cost reduction, and service level improvement.
performance across these considerations. This distribution To accomplish these requirements, many studies have been
model has demonstrated to be capable of working with developed over the past five years, whose findings can be
inventory control considering slow-moving products and categorized into three different approaches. The first approach
including configurations with shorter time series [28]. refers to the tools, protocols, and systems which allow retailers
to keep track of their inventory location as well as inventory
IRI is aimed to produce financial/operational profit by loss. Examples of tools are RFID, RAIN RFID, Bar Codes,
upgrading in-stock position monitored through technology- and systems based on previous devices like a Smart shelf
based inventory audits, in comparison to recurrent system. The second approach refers to algorithms and systems
replenishment of stores or holding extra stock. In order for this
focused on inventory optimization. Trends aiming to find the
technology to be profitable, retailers need to know how many optimization in inventory management include the Bayesian
items are available in every store. The exactitude of inventory Estimation Method, Threshold Accepting and Differential
records provided by this technology is critical to fill both Evolution algorithms, Logistic Information system, and
online and in-store orders [15]. Multi-Channel Distribution Center system. Finally, the third
A MDP increases the predicted total income of a group of approach focuses on determining the order quantity
products in a finite period. The objective of this model relies considering costs and inventory management. The main
on detecting the best markdown policy for every item. In this tendencies applied in retail industries are the IIS, EOQ, JRP,
model, a policy details a discount percentage which is set for VMI, OE Distribution, Threshold Accepting, and Differential
every item for all inventory levels each week. The MDP model Evolution algorithms, MDP, AUD and IQD policies, and
presents however, an inconvenient when it comes to more than Fuzzy Inventory Management method.
two items, it shows a problem with the interaction among
items becoming inconceivable an optimal solution [31]. IV. CONCLUSION
Retail companies have acquired significant importance
The AUD policy presents a discount rate available for each
within several countries due to their high economic
item bought. On the other hand, in the IQD policy, the reduce
contribution. Therefore, the need to analyze their KPIs
price rate is applied exclusively to extra units above the total
becomes highly significant, as well as their different systems,
amount of items over which the reduce price rate is given. Due
methodologies, and tools used within inventory management
to the material cost being normally the bigger spending in the
and optimization. From the aspects mentioned above, the main
manufacturing process, the policy was stretched in order to
trends in inventory management within companies were
work with final products and raw materials in periods of
defined.
scarcity and the newsvendor answers the clients demands
through the period while the finished product is completely Regarding KPIs, findings reveal 22 important indicators
sold [19]. within inventory management that must be considered when
retailers evaluate their stock. Among them, ten primary
Fuzzy Inventory Management is presented as an option for
indicators were founded: inventory level, actual inventory and
traditional probabilistic methods which work with inventory
its relationship to the company's information system, shortage
management incertitude. In order to define an optimal EOQ,
or shortage frequency, frequency of product reordering or
the following variables are described as Fuzzy variables;
replenishment, service level, replacement frequency, product
backorder cost, holding cost, order cost, and demand. These
availability, inventory in excess, number of items on the shelf
variables are simulated between them by applying particle
and level of income or profit. These indicators allow the
swarm optimization and genetic algorithms. This model not
organization to know the state of the stock, to be managed
only reduces the total cost, the confidence that it is possible to
appropriately, and show an excellent service quality and
keep the total cost within the budget is also increased. Fuzzy
product availability image to the customer. The importance of
logic can be applied to other situations beyond determining
evaluating an inventory management system using indicators
periodic and continuous inventory policies [48].
is reflected in the main advantages, i.e., the decrease in
3) What are the current trends in inventory handling and monetary loss, higher operating performance, and a higher
management in retail companies? profit rate.
To define the current trends for inventory management in Overall, the evidence from this study suggests that order
the retail industry, it is convenient to consider critical aspects quantity, inventory localization, and optimization are the main
that allow knowing the effectiveness of the proposed models factors in which the systems, methodologies, and tools are
employing the 22 KPIs presented above. These aspects are focused. In this context, RFID systems are the most employed
identified as key performance indicators of the inventory
tools in retail industries in terms of solving location issues [9] A. Fink, Conducting research literature reviews: From the internet to
because they are capable of keeping track of inventory and paper. Sage publications, 2019.
provide a high confidence level on inventory records. [10] A. Cooke, D. Smith, y A. Booth, “Beyond PICO: the SPIDER tool fo r
qualitative evidence synthesis”, Qualitative health research, vol. 22,
Likewise, in terms of order quantity, systems like the EOQ, núm. 10, pp. 1435–1443, 2012.
JRP, the AUD and IQD policies, and MDP focus on [11] A. Fink, Conducting Research Literature Reviews: From the Internet
determining the correct order of items to accomplish to Paper, vol. 1, núm. 4. 2014.
optimization levels. Indeed, some studies showed that retailers [12] B. Kitchenham, “Guidelines for performing systematic literature
are working with VMI. In this methodology, the supplier reviews in software engineering”, Proceedings - International
controls the inventory according to the retailer behavior, Conference on Software Engineering, vol. 1, p. 65, 2007, doi:
leading a complete optimization of the SC. Furthermore, the 10.1145/1134285.1134500.
current research came up with algorithms focused on the [13] W. Zhou y S. Piramuthu, “Effects of ticket-switching on inventory
optimization of the inventory like the Bayesian Estimation management: Actual vs. information system-based data”, Decisio n
Support Systems, vol. 77, pp. 31–40, sep. 2015, doi:
Method, the LIS for Omni-Chanel, the Threshold and 10.1016/j.dss.2015.05.010.
Differential Algorithms, and Multi Chanel Distribution [14] Ö. Turgut, F. Taube, y S. Minner, “Data-driven retail inventory
Center. All these optimization algorithms and methodologies management with backroom effect”, OR Spectrum, vol. 40, núm. 4, pp.
mention the importance of having an integrated information 945–968, oct. 2018, doi: 10.1007/s00291-018-0511-9.
system that allows companies to perform their decision [15] R. Ishfaq y U. Raja, “Empirical evaluation of IRI mitigation strategies
process. in retail stores”, Journal of the Operational Research Society, pp. 1–14,
ago. 2019, doi: 10.1080/01605682.2019.1640592.
It is important to mention that all retailers may not be able [16] M. Barratt, T. J. Kull, y A. C. Sodero, “Inventory record inaccuracy
to employ these technologies due to their high cost of dynamics and the role of employees within multi-channel distribution
implementation and maintenance. To all those retailers with center inventory systems”, Journal of Operations Management, vol. 63,
limited resources, cheaper software is accessible that could núm. 1, pp. 6–24, nov. 2018, doi: 10.1016/j.jo m.2018.09.003.
help with the management of their inventory like bar codes or [17] R. Cui, D. J. Zhang, y A. Bassamboo, “Learning from Inventory
Availability Information: Evidence from Field Experiments o n
policies as EOQ, AUD, and IQD, which will allow optimizing Amazon”, Management Science, vol. 65, núm. 3, pp. 1216–1235, mar.
their stock without making considerable investments. 2019, doi: 10.1287/ mnsc.2017.2950.
[18] W. Zhang y K. Rajaram, “Managing limited retail space for basic
ACKNOWLEDGEM ENTS products: Space sharing vs. space dedication”, European Journal of
This study is part of the research project “Modelo de Operational Research, vol. 263, núm. 3, pp. 768–781, dic. 2017, doi:
10.1016/j.ejor.2017.05.045.
optimización para la gestión de diseño, inventarios y logística
de las tiendas en el sector minorista (retail) del Azuay” [19] M. Keramatpour, S. T. A. Niaki, y S. H. R. Pasandideh, “A bi-objectiv e
two-level newsvendor problem with discount policies and budget
supported by the Research Department of the University of constraint”, Computers & Industrial Engineering, vol. 120, pp. 192–
Cuenca (DIUC). The authors gratefully acknowledge the 205, jun. 2018, doi: 10.1016/j.cie.2018.04.040.
contributions and feedback provided by the IMAGINE Project [20] B. Turan, S. Minner, y R. F. Hartl, “A VNS approach to multi-locat io n
team. inventory redistribution with vehicle routing”, Computers &
Operations Research, vol. 78, pp. 526–536, feb. 2017, doi:
REFERENCES 10.1016/j.cor.2016.02.018.
[1] R. Ishfaq, C. C. Defee, B. J. Gibson, y U. Raja, “Realignment of the [21] J. G. Wilson y C. K. Anderson, “Joint Inventory and Pricin g
physical distribution process in omni-channel fulfillmen t ”, Decisions”, IFAC-PapersOnLine, vol. 48, núm. 3, pp. 238–241, 2015,
International Journal of Physical Distribution & Logistics doi: 10.1016/j.ifacol.2015.06.087.
Management, vol. 46, núm. 6/7, pp. 543–561, jul. 2016, doi: [22] R. Ishfaq, U. Raja, y S. Rao, “Seller-induced scarcity and price -
10.1108/IJPDLM-02-2015-0032. leadership: Impact on product returns in the internet retail supply
[2] J. Kembro y A. Norrman, “Exploring trends, implications and chain”, The International Journal of Logistics Management, vol. 27,
challenges for logistics information systems in omni-channels : núm. 2, pp. 552–569, ago. 2016, doi: 10.1108/IJLM-05-2014-0073.
Swedish retailers’ perception”, International Journal of Retail and [23] F. Thiesse y T. Buckel, “A comparison of RFID-based shelf
Distribution Management, vol. 47, núm. 4, pp. 384–411, 2019, doi: replenishment policies in retail stores under suboptimal read rates”,
10.1108/IJRDM-07-2017-0141. International Journal of Production Economics, vol. 159, pp. 126– 136,
[3] G. Hançerlioğulları, A. Şen, y E. A. Aktunç, “Demand uncertainty and ene. 2015, doi: 10.1016/j.ijpe.2014.09.002.
inventory turnover performance: an empirical analysis of the US retail [24] L. B. Sabir y J. A. Farooquie, “Effect of Different Dimensions of
industry”, International Journal of Physical Distribution and Logistics Inventory Management of Fruits and Vegetables on Profitability of
Management, vol. 46, núm. 6–7, pp. 681–708, 2016, doi: Retail Stores: An Empirical Study”, Global Business Review, vol. 19,
10.1108/IJPDLM-12-2014-0303. núm. 1, pp. 99–110, feb. 2018, doi: 10.1177/0972150917713278.
[4] J. D. Sterman y G. Dogan, “‘I’m not hoarding, i’m just stocking up [25] A. Solti, M. Raffel, G. Romagnoli, y J. Mendling, “Misplaced product
before the hoarders get here.’: Behavioral causes of phantom ordering detection using sensor data without planograms”, Decision Support
in supply chains”, Journal of Operations Management, vol. 39, pp. 6– Systems, vol. 112, pp. 76–87, ago. 2018, doi:
22, 2015. 10.1016/j.dss.2018.06.006.
[5] Y. Wang, S. W. Wallace, B. Shen, y T.-M. Choi, “Service supply chain [26] N. Kasiri, “More insights into RFID-enabled changes in retail: A
management: A review of operational models”, European Journal of simulation model”, International Journal of RF Technologies, vol. 7,
Operational Research, vol. 247, núm. 3, pp. 685–698, 2015. núm. 4, pp. 229–248, sep. 2016, doi: 10.3233/RFT -161650.
[6] S. Mahar y P. D. Wright, “The value of postponing online fulfillmen t [27] C. Eroglu, B. D. Williams, y M. A. Waller, “Using the Pack-and-a-Half
decisions in multi-channel retail/e-tail organizations”, Computers & Rule to Eliminate Backroom Inventories in Retail Operations”, Journal
operations research, vol. 36, núm. 11, pp. 3061–3072, 2009. of Business Logistics, vol. 39, núm. 3, pp. 164–181, sep. 2018, doi:
[7] A. Hübner, A. Holzapfel, y H. Kuhn, “Operations management in 10.1111/jbl.12191.
multi-channel retailing: an exploratory study”, Operations [28] G. J. Hahn y A. Leucht, “Managing inventory systems of slow-movin g
Management Research, vol. 8, núm. 3–4, pp. 84–100, 2015. items”, International Journal of Production Economics, vol. 170, pp.
[8] A. Hübner, H. Kuhn, J. Wollenburg, y A. Trautrims, “From bricks -and - 543–550, dic. 2015, doi: 10.1016/j.ijpe.2015.08.014.
mortar to bricks-and-clicks–logistics networks in omni-channel [29] T.-M. Choi, “Inventory Service Target in Quick Response Fashion
grocery retailing”, Empirical Studies in Multi-Channel and Omn i- Retail Supply Chains”, Service Science, vol. 8, núm. 4, pp. 406– 419,
Channel Retail Operations and Logistics, p. 102, 2018. dic. 2016, doi: 10.1287/serv.2016.0146.
[30] D. Fan, Q. Xu, T. Fan, y F. Cheng, “Inventory optimization model motor disabilities in the context of Smart Cities”, Future Generatio n
considering consumer shift and inventory transshipment in dual- Computer Systems, vol. 76, pp. 248–261, 2017, doi:
channel supply chains”, RAIRO - Operations Research, vol. 53, núm. 10.1016/j.future.2016.11.030.
1, pp. 59–79, ene. 2019, doi: 10.1051/ro/2018045. [41] H. H. C. Chuang, “Mathematical modeling and Bayesian estimation for
[31] Ö. Cosgun, U. Kula, y C. Kahraman, “Markdown optimization for an error-prone retail shelf audits”, Decision Support Systems, vol. 80, pp.
apparel retailer under cross -price and initial inventory effects”, 72–82, 2015, doi: 10.1016/j.dss.2015.10.003.
Knowledge-Based Systems, vol. 120, pp. 186–197, mar. 2017, doi: [42] A. Mahmoodi, “Stackelberg–Nash equilibrium of pricing and
10.1016/j.knosys.2017.01.003. inventory decisions in duopoly supply chains using a nested
[32] L.-H. Zhang, T. Li, y T.-J. Fan, “Radio-frequency identificatio n evolutionary algorithm”, Applied Soft Computing Journal, vol. 86,
(RFID) adoption with inventory misplacement under retail núm. xxxx, p. 105922, 2020, doi: 10.1016/j.asoc.2019.105922.
competition”, European Journal of Operational Research, vol. 270, [43] A. Hübner, J. Wollenburg, y A. Holzapfel, “Retail logistics in the
núm. 3, pp. 1028–1043, nov. 2018, doi: 10.1016/j.ejor.2018.04.038. transition from multi-channel to omni-channel”, International Journal
[33] W. Qin, R. Y. Zhong, H. Y. Dai, y Z. L. Zhuang, “An assessment model of Physical Distribution and Logistics Management, vol. 46, núm. 6–7,
for RFID impacts on prevention and visibility of inventory inaccuracy pp. 562–583, 2016, doi: 10.1108/IJPDLM-08-2015-0179.
presence”, Advanced Engineering Informatics, vol. 34, pp. 70–79, oct. [44] S. Tiwari, L. E. Cárdenas-Barrón, A. A. Shaikh, y M. Goh, “Retailer’s
2017, doi: 10.1016/j.aei.2017.09.006. optimal ordering policy for deteriorating items under order-size
[34] C. I. Papanagnou y O. Matthews -Amune, “Coping with demand dependent trade credit and complete backlogging”, Computers and
volatility in retail pharmacies with the aid of big data exploration ”, Industrial Engineering, vol. 139, 2020, doi: 10.1016/j.cie.2018.12 .0 0 6.
Computers & Operations Research, vol. 98, pp. 343–354, oct. 2018, [45] A. Ross, M. Khajehnezhad, W. Otieno, y O. Aydas, “Integrated
doi: 10.1016/j.cor.2017.08.009. location-inventory modelling under forward and reverse product flows
[35] M. Bieniek, “The ubiquitous nature of inventory: Vendor Managed in the used merchandise retail sector: A multi-echelon formulation ”,
Consignment Inventory in adverse market conditions”, European European Journal of Operational Research, vol. 259, núm. 2, pp. 664 –
Journal of Operational Research, ago. 2019, doi: 676, 2017, doi: 10.1016/j.ejor.2016.10.036.
10.1016/j.ejor.2019.07.070. [46] N. K. Verma y A. K. Chatterjee, “A multiple-retailer replenishmen t
[36] A. Rizzi y A. Volpi, “RFID-enabled visual merchandising in apparel model under VMI: Accounting for the retailer heterogeneity”,
retail”, International Journal of RF Technologies: Research and Computers and Industrial Engineering, vol. 104, pp. 175–187, 2017,
Applications, vol. 8, núm. 4, pp. 213–231, 2018, doi: 10.3233/ RFT - doi: 10.1016/j.cie.2016.12.001.
181788. [47] B. S. Onggo, J. Panadero, C. G. Corlu, y A. A. Juan, “Agri-food supply
[37] H. Farhat, P. Iliev, P. Marriage, y N. Rolland, “An added value chains with stochastic demands: A multi-period inventory routing
alternative to RAIN RFID items characterization in retail”, IEEE problem with perishable products”, Simulation Modelling Practice and
Access, vol. 6, pp. 32430–32439, 2018, doi: Theory, vol. 97, núm. May, p. 101970, 2019, doi:
10.1109/ACCESS.2018.2844739. 10.1016/j.simpat.2019.101970.
[38] J. R. García Oya, R. Martín Clemente, E. Hidalgo Fort, R. González [48] P. Wanke, H. Alvarenga, H. Correa, A. Hadi-Vencheh, y M. A. K.
Carvajal, y F. Muñoz Chavero, “Passive RFID-Based Inventory of Azad, “Fuzzy inference systems and inventory allocation decisions:
Traffic Signs on Roads and Urban Environments”, Sensors (Basel, Exploring the impact of priority rules on total costs and service levels”,
Switzerland), vol. 18, núm. 7, 2018, doi: 10.3390/s18072385. Expert Systems with Applications, vol. 85, pp. 182–193, 2017, doi:
[39] X. Liu, K. Li, G. Min, Y. Shen, A. X. Liu, y W. Qu, “Completely 10.1016/j.eswa.2017.05.043.
pinpointing the missing RFID tags in a time-efficient way”, IEEE
Transactions on Computers, vol. 64, núm. 1, pp. 87–96, 2015, doi:
10.1109/TC.2013.197.
[40] Z. Rashid, J. Melià-Seguí, R. Pous, y E. Peig, “Using Augmented
Reality and Internet of Things to improve accessibility of people wit h

View publication stats