KR102508232B1 - Method, device and system for managing distribution of inventory for mart operation based on artificial intelligence - Google Patents

Abstract

According to one embodiment of the present invention, a method for managing distribution of inventory for a mart operation based on artificial intelligence, performed by a device, comprises: a step of analyzing a purchase pattern of purchasing first products in a first mart based on purchase history of the first mart; a step of applying an analysis result of the purchase pattern of the first products to a first artificial neural network and setting a first period, which is an inventory update period of the first products, based on output of the first artificial neural network; a step of classifying the first products, stored in the first mart, by expiration date every first period to classify products, having the shortest remaining period until the expiration date from among the first products, into a first group; a step of confirming the amount of inventory of the first products, classified into the first group, as first inventory and confirming a remaining period until the expiration date of the first products, classified into the first group, as a first period; and a step of sending a notification message, notifying that the expiration date for the amount of the first inventory in the inventory of the first products is imminent, when it is confirmed that the first period is shorter than a pre-set first reference period. Accordingly, the present invention has an effect of facilitating the management of expiration dates of products sold in a market.

Description

Inventory distribution management method, device and system for artificial intelligence-based mart operation {METHOD, DEVICE AND SYSTEM FOR MANAGING DISTRIBUTION OF INVENTORY FOR MART OPERATION BASED ON ARTIFICIAL INTELLIGENCE}

The following embodiments relate to technologies for managing the distribution of inventory for mart operation based on artificial intelligence.

Products sold in marts have a fixed expiration date, so if they are not sold within the expiration date, the product has no choice but to be discarded.

In addition, since the expiration date is set differently according to the production time even for the same product, it is difficult to manage the expiration date of the products.

Therefore, there is an increasing demand to manage the distribution of inventory by classifying and managing each product by expiration date, providing a notification service for the expiration date of product inventory according to the deadline, and research on related technologies is required. .

Korean Patent Registration No. 10-1743353 Korean Registered Patent No. 10-2057001 Korean Patent Registration No. 10-2327747 Korean Patent Registration No. 10-2337279

According to an embodiment, based on the purchase history of the first mart, the purchase pattern of purchasing the first product in the first mart is analyzed, and the analysis result of the purchase pattern of the first product is applied to the first artificial neural network. , Based on the output of the first artificial neural network, the first cycle, which is the inventory update cycle of the first product, is set, and the first product stored in the first mart is classified by expiration date for each first cycle, and the first product Among them, products with the shortest remaining period until expiration date are classified as a first group, the stock amount of the first product classified as the first group is confirmed as the first stock amount, and the first product classified as the first group is distributed. A notification message notifying that the expiration date for the first stock amount among the inventory of the first product is imminent when the remaining period remaining until the due date is determined as the first period, and the first period is determined to be shorter than the preset first reference period. Its purpose is to provide a method, device, and system for managing the distribution of inventory for artificial intelligence-based mart operation, which transmits to the first mart terminal.

The object of the present invention is not limited to the object mentioned above, and other objects not mentioned will be clearly understood from the description below.

According to an embodiment, in the stock distribution management method for operating an artificial intelligence-based mart, performed by an apparatus, a purchase pattern of purchasing a first product from the first mart based on purchase details of the first mart analyzing the; setting a first period, which is a stock update period of the first product, based on an output of the first artificial neural network by applying an analysis result of the purchase pattern of the first product to a first artificial neural network; classifying the first products stored in the first mart by expiration date for each first cycle, and classifying products having the shortest remaining period until expiration date among the first products into a first group; identifying a stock amount of the first product classified into the first group as a first stock amount, and determining a remaining period until an expiration date of the first product classified into the first group as a first period; And if it is determined that the first period is shorter than the preset first reference period, transmitting a notification message informing that the expiration date for the first stock amount of the first product in stock is imminent, to the first mart terminal. Including, a distribution management method of inventory for artificial intelligence-based mart operation is provided.

In the step of transmitting a notification message informing that the expiration date for the first inventory amount is imminent to the first mart terminal, the daily average sales volume of the first product is determined as the first sales volume based on the sales details of the first mart. Confirming with; generating a first matching result by matching the first inventory amount, the first period, and the first sales volume; By applying the first matching result to the second artificial neural network, based on the output of the second artificial neural network, the probability that the first inventory amount remains unsold until the first period elapses is determined by first predicting by incidence; setting a distribution management state for the first stock amount to a warning state when it is confirmed that the first occurrence rate is equal to or greater than a preset first reference value; setting a distribution management state for the first inventory amount to a caution state when it is determined that the first occurrence rate is less than the first reference value and equal to or greater than a preset second reference value; setting a distribution management state for a first stock amount to a normal state when it is confirmed that the first occurrence rate is less than the second reference value; Transmitting a notification message recommending a discount for the first product to the first mart terminal when the distribution management state for the first stock amount is set to a warning state; transmitting a notification message recommending a promotion for the first product to the first mart terminal when the distribution management state for the first stock amount is set to a caution state; and transmitting a notification message recommending purchase of the first product to the first mart terminal when the distribution management state for the first inventory amount is set to a normal state.

Transmitting a notification message recommending a discount for the first product to the first mart terminal may include setting a first ratio to 99% when it is determined that the first inventory amount is less than a preset first reference amount; setting the first ratio to a value lower than 99% as the first inventory amount increases when it is determined that the first stock amount is equal to or greater than the first reference amount; setting the second ratio to 99% when it is determined that the first period is longer than a preset second reference period; if it is determined that the first period is not longer than the second reference period, setting the second ratio to a value lower than 99% as the first period is shorter; calculating a second price by multiplying the first price, the first ratio, and the second ratio when the price of the first product is set as the first price; and changing the price of the first product from the first price to the second price, and transmitting a notification message recommending a discount for the first product to the first mart terminal.

According to an embodiment, each product is classified and managed by expiration date, and a notification service is provided to notify the expiration date of product stock according to the deadline, thereby helping to easily manage the expiration date of products sold in a mart. It works.

On the other hand, the effects according to the embodiments are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a diagram schematically showing the configuration of a system according to an embodiment.
2 is a flowchart illustrating a process of performing inventory distribution management for mart operation based on artificial intelligence according to an embodiment.
3 is a flowchart illustrating a process of setting a distribution management state of a product by estimating a probability that a product's stock quantity remains unsold, according to an embodiment.
4 is a flowchart illustrating a process of providing a notification service recommending a discount at an optimized price according to an inventory amount and expiration date according to an embodiment.
5 is a flowchart illustrating a process of providing a notification service recommending an optimized purchase amount according to an inventory amount and expiration date according to an embodiment.
6 is a flowchart illustrating a process of providing rewards according to an embodiment.
7 is a flowchart illustrating a process of calculating a performance rate according to an exemplary embodiment.
8 is a diagram for explaining learning of a first artificial neural network according to an embodiment.
9 is a diagram for explaining learning of a second artificial neural network according to an embodiment.
10 is an exemplary diagram of a configuration of a device according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only, and may be modified and implemented in various forms. Therefore, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Although terms such as first or second may be used to describe various components, such terms should only be construed for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element.

It should be understood that when an element is referred to as being “connected” to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle.

Terms used in the examples are used only for descriptive purposes and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

The embodiments may be implemented in various types of products such as personal computers, laptop computers, tablet computers, smart phones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices.

In an embodiment, an artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, a machine learns and judges on its own. The more the AI system is used, the higher the recognition rate and the more accurately the seller's taste can be understood, so the existing rule-based smart system is gradually being replaced by a deep learning-based AI system.

Artificial intelligence technology consists of machine learning and element technologies using machine learning. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and element technology is a technology that uses machine learning algorithms such as deep learning to mimic functions such as recognition and judgment of the human brain. It consists of technical fields such as understanding, inference/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology for recognizing and applying/processing human language/characters, including natural language processing, machine translation, dialogue systems, question and answering, voice recognition/synthesis, and the like. Visual understanding is a technology for recognizing and processing objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, space understanding, image improvement, and the like. Inference prediction is a technique of reasoning and predicting logically by judging information, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification) and knowledge management (data utilization). Motion control is a technology for controlling the autonomous driving of a vehicle and the movement of a robot, and includes motion control (navigation, collision, driving), manipulation control (action control), and the like.

In general, in order to apply machine learning algorithms to real life, learning is performed in a trial and error method due to the nature of the basic methodology of machine learning. In particular, deep learning requires hundreds of thousands of iterations. It is impossible to execute this in an actual physical external environment, so instead, the actual physical external environment is virtually implemented on a computer and learning is performed through simulation.

1 is a diagram schematically showing the configuration of a system according to an embodiment.

Referring to FIG. 1 , a system according to an embodiment may include a plurality of mart terminals 100 and a device 200 capable of communicating with each other through a communication network.

First, a communication network may be configured regardless of its communication mode, such as wired or wireless, and may be implemented in various forms so that communication between servers and communication between servers and terminals is performed.

Each of the plurality of mart terminals 100 may be implemented as a computing device having a communication function, and may be implemented as, for example, a mobile phone, a desktop PC, a laptop PC, a tablet PC, a smartphone, a POS terminal, etc. It is not limited and may be implemented in various types of communication devices that can be connected to external servers.

The plurality of mart terminals 100 are terminals used in marts that sell products, and may include a first mart terminal 110 used in a first mart, a second mart terminal 120 used in a second mart, and the like can

Hereinafter, for convenience of description, the operation of the first mart terminal 110 will be mainly described, but other mart terminals such as the second mart terminal 120 can perform the operation of the first mart terminal 110 instead, of course am.

Each of the plurality of mart terminals 100 may access the device 200 through a web page, application, or the like provided by the device 200 . To this end, each of the plurality of mart terminals 100 may be installed with an application providing an inventory distribution management service for mart operation. The corresponding application may operate in conjunction with the device 200 .

The device 200 may be a server owned by a person or organization that provides services using the device 200, may be a cloud server, or may be a peer-to-peer (p2p) set of distributed nodes. may be The device 200 may be configured to perform all or some of the arithmetic function, storage/reference function, input/output function, and control function of a normal computer. The device 200 may include one or more artificial neural networks that perform an inference function.

The device 200 may be configured to communicate with a plurality of mart terminals 100 by wire or wirelessly, control the operation of each of the plurality of mart terminals 100, and display certain information on the screen of each of the plurality of mart terminals 100. You can control what is displayed.

The device 200 may be implemented as a server that provides a service for managing distribution of inventory for operation of a mart, and may provide a platform service for managing distribution of inventory for operation of a mart.

Meanwhile, for convenience of description, FIG. 1 shows only the first mart terminal 110 and the second mart terminal 120 among the plurality of mart terminals 100, but the number of terminals may vary according to the embodiment. . As long as the processing capacity of the device 200 allows, the number of terminals is not particularly limited.

The device 200 may provide a notification service of the expiration date of products purchased at the mart according to the expiration date, and may optimize the price to increase the sales rate of the product with an imminent expiration date.

According to an embodiment, the device 200 may set a product inventory update cycle based on artificial intelligence.

In addition, the device 200 may predict the probability that the stock of a product remains unsold based on artificial intelligence.

In the present invention, artificial intelligence (AI) means a technology that imitates human learning ability, reasoning ability, perception ability, etc., and implements it with a computer, and concepts such as machine learning and symbolic logic can include Machine learning (ML) is an algorithm technology that classifies or learns the characteristics of input data by itself. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the result of the analysis, and can make judgments or predictions based on the result of the learning. In addition, technologies that mimic the functions of the human brain, such as recognition and judgment, using machine learning algorithms, can also be understood as artificial intelligence. For example, technical fields such as linguistic understanding, visual understanding, inference/prediction, knowledge expression, and motion control may be included.

Machine learning may refer to processing that trains a neural network model using experience of processing data. Through machine learning, computer software could mean improving its own data processing capabilities. A neural network model is constructed by modeling a correlation between data, and the correlation may be expressed by a plurality of parameters. The neural network model derives a correlation between data by extracting and analyzing features from given data, and optimizing the parameters of the neural network model by repeating this process can be referred to as machine learning. For example, a neural network model may learn a mapping (correlation) between an input and an output with respect to data given as an input/output pair. Alternatively, even when only input data is given, the neural network model may learn the relationship by deriving a regularity between given data.

An artificial intelligence learning model or a neural network model may be designed to implement a human brain structure on a computer, and may include a plurality of network nodes having weights while simulating neurons of a human neural network. A plurality of network nodes may have a connection relationship between them by simulating synaptic activities of neurons that transmit and receive signals through synapses. In the artificial intelligence learning model, a plurality of network nodes can send and receive data according to a convolutional connection relationship while being located in layers of different depths. The artificial intelligence learning model may be, for example, an artificial neural network model, a convolutional neural network model (CNN), and the like. As an embodiment, the artificial intelligence learning model may be machine-learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, Ada-boost , Perceptron, Genetic Programming, Clustering, etc. may be used.

Of these, CNNs are a type of multilayer perceptrons designed to use minimal preprocessing. A CNN consists of one or several convolution layers and general artificial neural network layers on top, and additionally utilizes weights and pooling layers. Thanks to this structure, CNN can fully utilize the input data with a two-dimensional structure. Compared to other deep learning structures, CNN shows good performance in both video and audio fields. CNNs can also be trained via standard back-propagation. CNNs are easier to train than other feedforward artificial neural network techniques and have the advantage of using fewer parameters.

Convolutional networks are neural networks that contain sets of nodes with bound parameters. The increasing size of available training data and the availability of computational power, combined with algorithmic advances such as piecewise linear unit and dropout training, have greatly improved many computer vision tasks. With huge data sets, such as those available for many tasks today, overfitting is not critical, and increasing the size of the network improves test accuracy. Optimal use of computing resources becomes a limiting factor. To this end, a distributed, scalable implementation of deep neural networks can be used.

2 is a flowchart illustrating a process of performing inventory distribution management for mart operation based on artificial intelligence according to an embodiment.

Referring to FIG. 2 , first, in step S201 , the device 200 may analyze a purchase pattern of purchasing a first product from the first mart based on purchase details of the first mart. Here, the purchase details are accumulated information of purchase details purchased by ordering products at the mart, and may be composed of a list indicating how many products were purchased at a certain point in time, and information such as product name, order date, warehousing date, purchase amount, etc. can include

According to one embodiment, the purchase details include information such as product name, order date, stock date, purchase amount, and after ordering a product at the mart, if it is confirmed that the ordered product has been received at the mart, the product is purchased at the mart. can be figured out That is, the warehousing date may be identified as the purchase date when the product is purchased at the mart.

Purchase details classified for each mart are stored in the database of the device 200, and the device 200 may obtain the purchase details of the first mart by inquiring information stored in the database.

For example, as a result of checking the purchase details of the first product in the first mart through the purchase history of the first mart, 20 items on January 1, 15 items on January 11, and 25 items on January 21 When it is confirmed that the purchase has been made, the device 200 may analyze the purchase pattern of the first product by identifying that the purchase of the first product is made by an average of 20 units at 10-day intervals. Here, the purchase pattern of the first product may refer to a purchase pattern indicating how often and in what quantity the first product is purchased at the first mart.

In step S202, the apparatus 200 may apply the analysis result of the embedding pattern of the first product to the pre-learned first artificial neural network.

According to an embodiment, the first artificial neural network may be an algorithm that outputs a product inventory update cycle suitable for the product purchase pattern after receiving an analysis result of the product purchase pattern. For example, if the purchase pattern of the first product is analyzed as a pattern in which purchase is made at 10-day intervals, the first artificial neural network analyzes the purchase pattern of the first product analyzed as being made at 10-day intervals. It is received as an input, and the inventory update cycle of products suitable for the purchase pattern made at 10-day intervals can be output as 10 days.

In step S203 , the device 200 may set a first cycle, which is a stock update cycle of the first product, based on the output of the first artificial neural network. That is, the first artificial neural network analyzes the purchase pattern of purchasing the first product in the first mart, and outputs a set value for the first cycle, which is a stock update cycle of the first product suitable for the purchase pattern of the first product. And, the device 200 may set the inventory update cycle of the first product to the first cycle based on the output of the first artificial neural network.

To this end, the first artificial neural network may be trained to select and extract a product inventory update cycle suitable for a product purchase pattern through an analysis result of a product purchase pattern. This first artificial neural network may be learned through a method described later with reference to FIG. 8 .

In step S204, the device 200 may classify the first product stored in the first mart for each first cycle by expiration date.

Specifically, the device 200 may classify the first product stored in the first mart for each first period by expiration date based on the inventory information of the first mart. Here, the inventory information of the first mart may include a stock status of each product stored in the first mart.

The device 200 may update inventory information of the first mart in real time by tracking new purchases of products in the first mart or sales of products in the first mart. To this end, since barcodes are attached to products handled in the first mart, tracking and management from purchase to sale can be performed through barcode recognition.

That is, when the device 200 is tracked and managed from purchase to sale of products through barcode recognition in each of a plurality of marts, it is possible to grasp the inventory status of products that change in real time for each mart through barcode recognition, and for each mart Separated inventory information can be stored and managed in a database.

According to an embodiment, the inventory information may include information such as product name, expiration date, stock quantity, price, and the like, and even if the same product has different expiration dates, it may be classified and managed according to expiration dates. Through this, the device 200 may classify and classify the first product by expiration date based on the inventory information of the first mart.

In step S205, as a result of classifying the first products stored in the first mart by expiration date, the device 200 may classify products having the shortest remaining period until the expiration date among the first products into the first group. there is.

For example, in the inventory information of Mart 1, “Product 1, expiration date January 20, stock amount 10 units”, “Product 1, expiration date January 25, inventory quantity 15 items”, “Product 1, expiration date January 25, inventory quantity 15 units” If information such as “expiration date January 30, stock amount 30 units” is included, the device 200 determines January 20 as the shortest remaining period until the expiration date among the first products, and January 20 Ten first products whose expiration dates are days can be classified into a first group.

In step S206, the device 200 determines the inventory amount of the first product classified in the first group as the first inventory amount, and determines the remaining period until the expiration date of the first product classified in the first group as the first period. You can check.

For example, if the device 200 classifies 10 first products whose expiration date is January 20 into the first group, determines the first stock quantity as 10, and determines the current date as January 15, The first period can be identified as 5 days.

In step S207, the device 200 may check whether the first period is shorter than the first reference period. Here, the first reference period may be set differently according to embodiments.

If it is confirmed in step S207 that the first period is not shorter than the first reference period, after a certain period of time, the process returns to step S201, and the device 200 may perform again from the process of analyzing the purchase pattern of the first product. there is.

For example, if the first reference period is 5 days and the first period is 8 days, the device 200 determines that the expiration date is not imminent because the first period is confirmed to be longer than the first reference period, It may be determined that it is not necessary to provide a notification service notifying that the expiration date is imminent.

If it is determined in step S207 that the first period is shorter than the first reference period, in step S208, the device 200 sends a notification message informing that the expiration date for the first inventory amount of the first product is imminent, to the first mart. It can be transmitted to the terminal 110. Here, the notification message may include information about the product name of the first product, the first stock amount, the expiration date of the first product classified into the first group, and the first period.

For example, when the first reference period is 5 days and the first period is 3 days, the device 200 determines that the expiration date is imminent because the first reference period is confirmed to be shorter than the first reference period, and It may be determined that it is necessary to provide a notification service notifying of the imminent occurrence, and through this, a notification message informing that the expiration date for the first stock amount of the first product in stock is imminent is transmitted to the first mart terminal 110. can

The device 200 transmits a notification message informing that the expiration date of the first stock amount of the first product is imminent to the first mart terminal 110, while recommending a discount, promotion, or purchase of the first product. A notification message may be further transmitted to the first mart terminal 110, and a detailed description related to this will be described later with reference to FIG. 3.

3 is a flowchart illustrating a process of setting a distribution management state of a product by estimating a probability that a product's stock quantity remains unsold, according to an embodiment.

Referring to FIG. 3 , first, in step S301, the device 200 may determine the daily average sales volume of the first product as the first sales volume based on the sales details of the first mart. Here, the sales details are accumulated information of sales details of products sold at the mart, and may consist of a list indicating how many products were sold at a certain point in time, and may include information such as product name, sale date, and amount sold. there is.

Sales details classified by mart are stored in the database of the device 200, and the device 200 can obtain the sales details of the first mart by inquiring information stored in the database.

For example, as a result of checking the purchase details of the first product in the first mart through the sales history of the first mart, 18 on January 1, 20 on January 2, and 22 on January 3 If it is confirmed that the purchase has been made, the device 200 may calculate the daily average sales volume of the first product as 20 units, and may identify 20 units as the first sales volume.

In step S302 , the device 200 may generate a first matching result by matching the first inventory amount, the first period, and the first sales volume.

In step S303, the device 200 may apply a first matching result obtained by matching the first inventory amount, the first period, and the first sales volume to the pretrained second artificial neural network. Here, the second artificial neural network receives as input a matching result of matching the inventory amount, the remaining period until the expiration date, and the daily average sales volume, and outputs a probability that the inventory amount remains unsold until the expiration date has elapsed. It can be an algorithm.

That is, the second artificial neural network may estimate and calculate the probability that the stock quantity remains unsold until the expiration date, in consideration of the stock quantity, the remaining period until the expiration date, and the daily average sales volume.

For example, if the first matching result of matching the first inventory amount, the first period, and the first sales volume is (60 pieces, 5 days, 10 pieces), the second artificial neural network is (60 pieces, 5 days, 10 pieces) ) as an input, the probability that 60 stocks remain unsold after 5 days can be predicted and calculated as the first occurrence rate, and an output value indicating the first occurrence rate can be output.

In step S304 , the apparatus 200 may predict a probability that the first stock quantity remains unsold until the first period elapses as the first occurrence rate, based on the output of the second artificial neural network.

According to an embodiment, the second artificial neural network determines the probability that the inventory amount remains unsold until the expiration date through a matching result of matching the inventory amount, the remaining period until the expiration date, and the daily average sales volume. You can analyze and print how much. The second artificial neural network may be learned through a method described below with reference to FIG. 9 .

In step S305, the device 200 may determine whether the first incidence rate is greater than or equal to a first reference value. Here, the first reference value may be set differently according to embodiments.

If it is determined in step S305 that the first occurrence rate is greater than or equal to the first reference value, in step S306, the device 200 may set the distribution management state for the first inventory amount to a warning state.

For example, when the first reference value is 60% and the first occurrence rate is confirmed to be 70%, the device 200 determines that the probability that the first inventory amount remains without being sold is very high, and the first inventory amount The distribution management status can be set to a warning status.

In step S307 , the device 200 may transmit a notification message recommending a discount for the first product to the first mart terminal 110 when the distribution management state for the first inventory amount is set to a warning state. Here, the notification message may include information about a discount price, a discount rate, a product price before discount, a product price after discount, and the like.

When the device 200 transmits a notification message recommending a discount for the first product to the first mart terminal 110, the device 200 optimizes the price of the first product to determine how much to discount, and recommends a discount at the optimized price. A notification message can be transmitted to the first mart terminal 110, and a detailed description related to this will be described later with reference to FIG. 4.

If it is determined in step S305 that the first incidence rate is less than the first reference value, in step S308, the device 200 may determine whether the first incidence rate is greater than or equal to the second reference value. Here, the second reference value may be set to a value smaller than the first reference value.

If it is determined in step S308 that the first occurrence rate is greater than or equal to the second reference value, in step S309 , the device 200 may set the distribution management state for the first inventory amount to a caution state.

For example, when the first reference value is 60% and the second reference value is 30%, the device 200 determines the probability that the first inventory amount remains unsold when the first occurrence rate is 40%. Upon determining that there is, a distribution management state for the first inventory amount may be set to a caution state.

In step S310, the device 200 may transmit a notification message recommending a promotion for the first product to the first mart terminal 110 when the distribution management state for the first inventory amount is set to a caution state. Here, the notification message may include information about an image to be used for a print advertisement, video content to be used for an online advertisement, and the like.

If it is determined in step S308 that the first occurrence rate is less than the second reference value, in step S311 , the device 200 may set the distribution management state for the first inventory amount to a normal state.

For example, when the first reference value is 60% and the second reference value is 30%, the apparatus 200 determines that the first occurrence rate is 20%, the probability that the first inventory amount remains unsold is very low. Upon determining that, the distribution management state for the first inventory amount may be set to a normal state.

In step S312, the device 200 may transmit a notification message recommending purchase of the first product to the first mart terminal 110 when the distribution management state for the first inventory amount is set to a normal state. Here, the notification message may include information about the purchase amount and the like.

When the device 200 transmits a notification message recommending purchase of the first product to the first mart terminal 110, the device 200 may transmit a notification message recommending the amount of purchase to the first mart terminal 110, indicating how much to purchase. And, a detailed description related to this will be described later with reference to FIG. 5 .

4 is a flowchart illustrating a process of providing a notification service recommending a discount at an optimized price according to an inventory amount and expiration date according to an embodiment.

Referring to FIG. 4 , first, in step S401, the device 200 may determine whether or not the first stock amount is less than the first reference amount. Here, the first reference amount may be set differently according to embodiments.

If it is confirmed in step S401 that the first stock amount is less than the first reference amount, in step S402, the device 200 may set the first ratio to 99%.

For example, when the first reference amount is 10, the device 200 may set the first ratio to 99% when the first stock amount is 9.

If it is determined in step S401 that the first stock amount is greater than or equal to the first reference amount, in step S403 , the device 200 may set the first ratio to a value lower than 99% as the first stock amount increases.

For example, when the first reference amount is 10, the device 200 sets the first ratio to 98% when the first stock amount is 10, and sets the first ratio to 97 when the first stock amount is 11. When set as % and the first stock amount is confirmed as 12, the first ratio may be set to 96%.

That is, the device 200 may set the first ratio to a lower value for a price discount due to an abundance of stock when the first stock amount is greater than a predetermined level.

Meanwhile, in step S404, the device 200 may check whether the first period is longer than the second reference period. Here, the second reference period may be set differently according to embodiments.

If it is confirmed in step S404 that the first period is longer than the second reference period, in step S405, the device 200 may set the second ratio to 99%.

For example, if the second reference period is 5 days and the first period is 6 days, the device 200 may set the second ratio to 99%.

If it is confirmed in step S404 that the first period is not longer than the second reference period, in step S406, the device 200 may set the second ratio to a value lower than 99% as the first period is shorter.

For example, when the second reference period is 5 days, the device 200 sets the second ratio to 98% when the first period is 5 days, and sets the second ratio to 98% when the first period is 4 days. If it is set to 97% and the first period is confirmed to be 3 days, the second ratio may be set to 96%.

That is, the device 200 may set the second ratio to a lower value for a price discount due to an imminent expiration date when the first period is shorter than a predetermined period.

In step S407, the device 200 may confirm that the price of the first product is set to the first price based on the stock information of the first mart. In this case, the first price may mean the price of the first product before discount. The first price may be automatically set by a supplier supplying the first product or manually set by a manager of the first mart.

In step S408, when the price of the first product is set as the first price, the second price may be calculated by multiplying the first price, the first ratio, and the second ratio. In this case, the second price may mean the price of the first product after discounting.

For example, when the first price is 10000 won, the device 200 sets the second price through “10000 × 0.95 × 0.90” when the first rate is set to 95% and the second rate is set to 90%. It can be calculated as 8550 won.

In step S409 , the device 200 may change the price of the first product from the first price to the second price and transmit a notification message recommending a discount for the first product to the first mart terminal 110 . Here, the notification message may include information about a first price, a second price, a discount price calculated through a comparison of the first price and the second price, and a discount rate.

5 is a flowchart illustrating a process of providing a notification service recommending an optimized purchase amount according to an inventory amount and expiration date according to an embodiment.

Referring to FIG. 5 , first, in step S501 , the apparatus 200 may classify, among the first products, products having a shorter remaining period until the expiration date, next to the first group, into a second group.

For example, in the inventory information of Mart 1, “Product 1, expiration date January 20, stock amount 10 units”, “Product 1, expiration date January 25, inventory quantity 15 items”, “Product 1, expiration date January 25, inventory quantity 15 units” If information such as “expiration date January 30, stock amount 30 units” is included, the device 200 determines January 20 as the shortest remaining period until the expiration date among the first products, and January 20 10 first products whose expiration dates are days can be classified into a first group, and among the first products, January 25 is identified as the second shortest remaining period until the expiration date, and January 25 is Fifteen first products with expiration dates may be classified into a second group.

In step S502, the device 200 determines the stock amount of the first product classified into the second group as the second stock amount, and determines the remaining period until the expiration date of the first product classified into the second group as the second period. You can check.

For example, when the device 200 classifies 15 first products whose expiration date is January 25 into the second group, determines the second stock quantity as 15 items, and determines that the current date is January 15, The second period can be identified as 10 days.

In step S503 , the device 200 may calculate the expected inventory amount of the first product by subtracting a value obtained by multiplying the first sales amount by the number of days of the second period from the sum of the first inventory amount and the second inventory amount.

For example, if the first inventory amount is 10, the second inventory amount is 15, the first sales amount is 5, and the second period is 4 days, the device 200 displays “(10 + 15) - (5 × 4) ”, it is possible to calculate the estimated stock amount of the first product as 5 pieces.

In step S504 , the device 200 may determine whether the expected stock amount of the first product is less than the standard stock amount of the first product. Here, the standard stock amount of the first product may be set differently according to embodiments.

If it is confirmed in step S504 that the expected stock amount of the first product is less than the standard stock amount of the first product, in step S505, the device 200 subtracts the expected stock amount of the first product from the standard stock amount of the first product, The first purchase amount can be calculated.

For example, when the expected stock amount of the first product is 3 and the standard stock amount of the first product is 5, the device 200 may calculate the first purchase amount as 2 through “5 - 3”.

In addition, when the expected stock amount of the first product is -2 and the standard stock amount of the first product is 5, the device 200 may calculate the first purchase amount as 7 through “5 - (-2)”. .

In step S506, the device 200 may transmit a notification message recommending purchase of the first purchase amount to the first mart terminal 110. Here, the notification message may include information about the expected stock amount of the first product, the standard stock amount of the first product, the first purchase amount, and the expected stock amount after purchase obtained by summing the expected stock amount and the first purchase amount of the first product.

According to one embodiment, a plurality of marts may be contracted with a company that operates the device 200 as a franchise store, and the device 200 directly participates in managing the affiliated store in each of the plurality of franchise stores, and according to the participation It is possible to provide a platform service that allows users to receive compensation.

The device 200 may transmit a plurality of notices related to franchise affiliated stores to each of the plurality of mart terminals 100 . For example, the device 200 may transmit a plurality of notices related to franchise affiliated stores to the first mart terminal 110 . Here, the notice is a notice for operating a franchise member store, and may include, for example, a price increase notice, a new menu release notice, a delivery guide notice, and the like.

The device 200 may receive announcement performance completion information about announcements completed in each of a plurality of mart terminals 100 among a plurality of announcements, from each of a plurality of mart terminals 100 . For example, the device 200 may receive announcement execution completion information about announcements completed in the first mart terminal 110 among a plurality of announcements from the first mart terminal 110 .

The apparatus 200 may calculate a performance rate for announcements completed in each of the plurality of mart terminals 100 based on the announcement performance completion information received from each of the plurality of mart terminals 100 . For example, based on the announcement execution completion information received from the first mart terminal 110, the performance rate for announcements completed in the first mart terminal 110 may be calculated. A detailed method of calculating the performance rate will be described later with reference to FIGS. 6 and 7 .

The apparatus 200 may determine a reward to be provided to each of the plurality of mart terminals 100 based on the performance rate of each of the plurality of mart terminals 100. At this time, the device 200 may determine a reward in direct proportion to the performance rate.

The reward may be a point usable in the franchise affiliate management platform, or may be a cryptocurrency (or virtual currency, coin). For example, the device 200 blocks the notification execution completion information to the node of the first mart terminal 110 in a blockchain network formed by using each of the plurality of mart terminals 100 as nodes. can be saved in the form In this case, the reward may be a cryptocurrency that is mined based on the announcement execution completion information and transferred to a cryptocurrency wallet allocated to the first mart terminal 110 .

Blockchain is a ledger management technology in which small-scale data called blocks are stored in a chain-type link-based distributed data storage environment created based on a peer-to-peer (P2P) method. It is a ledger management technology based on distributed computing technology that can view the results of changes in This is essentially a form of distributed data storage technology, which is a change list recorded in all participating nodes of continuously changing data, and can be devised to prevent arbitrary manipulation by operators of distributed nodes. Blockchain technology provides most cryptocurrencies, including Bitcoin, as compensation, and since the cryptocurrency transaction process is used in a decentralized electronic ledger, a server is operated on each computer of many users who run blockchain software, and a centralized Free transactions between individuals may be possible without an existing bank.

The apparatus 200 may guide each of the plurality of marts to manage the franchise member stores by transmitting various checklists for operating the franchise member stores to the plurality of mart terminals 100 .

The device 200 may transmit a device setting checklist, a logistics setting checklist, and a mart management checklist for operating a franchise member store to each of the plurality of mart terminals 100 .

The device setting checklist may be a checklist for inspecting and managing devices used to operate a franchise member store. The device setting checklist includes a plurality of check items for checking and managing, for example, a chest freezer, a stand freezer, a blender, a point-of-sales device, and the like. It may be a checklist that includes

The logistics setting checklist may be a checklist for checking and managing the setting (or setting) states of logistics necessary to operate a franchise member store. For example, it may be a checklist including a plurality of check items for inspecting and managing a logistics setting checklist and delivery or logistics arrangement status.

The mart management checklist may be a checklist for managing an operating state of a franchise member store. For example, the mart management checklist includes a plurality of check items for inspecting and managing product quality, temperature and cleanliness of equipment, raw material status, personal hygiene status, sanitary facility status, safety status, mart environment status, etc. It could be a checklist.

A manager of each of a plurality of marts may evaluate at least one checklist among a device setting checklist, a logistics setting checklist, and a mart management checklist through each of the plurality of mart terminals 100 . Each of the plurality of mart terminals 100 may transmit, to the device 200 , evaluation completion checklist information about at least one evaluation completed checklist.

The device 200 may determine a reward to be provided to each of the plurality of mart terminals 100 according to the evaluation completion checklist information received from each of the plurality of mart terminals 100 . The device 200 may determine a higher (more) reward to provide as the number of checklists that have been evaluated increases.

The reward may be a point usable in the mart participation type franchise member store management platform, or may be a cryptocurrency (or virtual currency, coin). For example, the device 200 stores evaluation completion checklist information in a block form in the node of the first mart terminal 110 in a blockchain network formed by using each of the plurality of mart terminals 100 as nodes. can In this case, the reward may be a cryptocurrency that is mined based on the evaluation completion checklist information and transferred to a cryptocurrency wallet allocated to the first mart terminal 110 .

That is, the device 200 transmits notice execution completion information and evaluation completion checklist information to the node of the first mart terminal 110 in a block chain network formed by using each of the plurality of mart terminals 100 as nodes. can be saved In this case, the rewards may be mined based on the announcement execution completion information and the evaluation completion checklist information and provided in cryptocurrency delivered to a cryptocurrency wallet allocated to the first mart terminal 110 .

6 is a flowchart illustrating a process of providing rewards according to an embodiment.

Referring to FIG. 6 , first, in step S601, when a plurality of marts are contracted as franchisees, the device 200 may transmit a plurality of announcements related to franchisees to the plurality of mart terminals 100. Here, the notice may be a notice for operating a franchise member store. The plurality of notices may include at least one mandatory notice and at least one optional notice.

In step S602 , the device 200 may receive, from the first mart terminal 110 , announcement execution completion information regarding announcements completed in the first mart terminal 110 among a plurality of announcements. At this time, the manager of the first mart may perform at least one announcement among a plurality of announcements through the first mart terminal 110 .

For example, if there are a first notice, a second notice, and a third notice as a plurality of notices, and the first mart performs only the first notice and the second notice, the first mart terminal ( 110) The information on the completion of announcement completion may be information indicating that the first announcement and the second announcement have been performed, but the third announcement has not been performed.

In step S603, the device 200 may calculate a performance rate for the announcement completed in the first mart terminal 110 based on the announcement execution completion information.

For example, if the announcement execution completion information that has been performed in the first mart terminal 110 is information indicating that the first announcement and the second announcement have been performed and the third announcement has not been performed, the device (200) may calculate the execution rate for the announcement completed in the first mart terminal 110 as (1/3) X 100 = 33.3%.

In step S604, the device 200 may transmit a reward determined according to the performance rate to the first mart terminal 110. Here, the reward may be determined in direct proportion to the performance rate.

For example, the device 200 may determine the reward as 1000 won when the performance rate is 100%, determine the reward as 500 won when the performance rate is 50%, and determine the reward as 500 won when the performance rate is 0%. It can be determined as 0 won. At this time, the reward may be a point usable in the mart participating franchise member store management platform or may be a cryptocurrency.

In step S605, the device 200 may transmit a device setting checklist, a distribution setting checklist, and a mart management checklist for operating a franchise member store to the first mart terminal 110. The device 200 may transmit, to the first mart terminal 110 , a device setting checklist, a logistics setting checklist, and a mart management checklist that can be directly evaluated by the first mart. The above-described checklists may be transmitted to the first mart terminal 110 at regular intervals (eg, one day, one week).

In step S606, the device 200 receives, from the first mart terminal 110, evaluation completion checklist information on at least one evaluated checklist among the equipment setting checklist, the logistics setting checklist, and the mart management checklist. can At this time, the manager of the first mart may evaluate at least one checklist among a device setting checklist, a logistics setting checklist, and a mart management checklist received through the first mart terminal 110 . The first mart terminal 110 may transmit evaluation completion checklist information about at least one evaluation completed checklist to the device 200 .

For example, when only the mart management checklist has been evaluated in the first mart terminal 110 and the equipment setting checklist and logistics setting checklist have not been evaluated, the evaluation completion checklist information includes only the evaluated mart management checklist can do.

In step S607, the device 200 may transmit a reward determined according to the evaluation completion checklist information to the first mart terminal 110. Here, the reward may be determined in direct proportion to the number of evaluated checklists.

In step S608, the device 200 may receive an order list from the first mart terminal 110. Here, the order list is a list for requesting items necessary for the first franchise mart, and the manager of the first mart creates the order list through the first mart terminal 110, and the generated order list can be sent to

In step S609, the device 200 responds to the request to use the rewards from the first mart terminal 110, and sends the order amount obtained by subtracting the rewards from the prices of items included in the order list to the first mart terminal 110. can transmit In this case, the manager of the first mart may select to use the rewards obtained during creation and transmission of the order list through the first mart terminal 110 . In this case, the device 200 may determine the order amount by subtracting the rewards from the prices of items included in the order list in response to the request to use the rewards from the first mart terminal 110 .

For example, when the total price of items included in the order list is 1 million won and the reward obtained from the first mart terminal 110 is 50,000 won, the device 200 may determine the order amount as 950,000 won. The device 200 may transmit the determined final order amount of 950,000 won to the first mart terminal 110 . In a similar way, the manager of the mart can freely use the acquired reward in the mart participating franchise member store management platform.

7 is a flowchart illustrating a process of calculating a performance rate according to an exemplary embodiment.

Referring to FIG. 7, first, in step S701, the device 200 performs at least one mandatory announcement and at least one selected announcement, respectively, based on the announcement execution completion information received from the first mart terminal 110. It is possible to check whether the execution of Here, the plurality of notices may include at least one mandatory notice and at least one optional notice.

In step S702, the device 200 may check whether at least one mandatory announcement has been performed.

For example, if there are 3 required announcements, the device 200 may check whether all 3 required announcements have been completed in the first mart terminal 110 .

If it is determined in step S702 that all of the at least one essential announcement has not been performed, in step S704, the device 200 may calculate the execution rate for the announcement completed in the first mart terminal 110 as 0%. That is, when all required announcements are not performed, it may be determined not to provide a reward by calculating the performance rate as 0%.

In step S705, the device 200 determines that there is a problem with the first franchise affiliated store corresponding to the first mart terminal 110 when at least one essential announcement is not all performed, and the first mart terminal 110 ), a confirmation notification message for at least one mandatory announcement may be transmitted.

When it is confirmed that at least one required announcement has been all performed in step S702, in step S703, the device 200 performs a plurality of announcements based on the number of announcements and the number of announcements completed in the first mart terminal 110. , It is possible to calculate an execution rate for announcements that have been completed in the first mart terminal 110 .

For example, if the number of the plurality of announcements includes 2 mandatory announcements and 3 optional announcements, and the first mart terminal 110 performs two mandatory announcements and one optional announcement, the device (200) can be calculated as a performance rate of 60% according to the formula below.

In Equation 1, X _n is the performance rate for announcements completed in a specific mart terminal, A _n is the number of announcements completed in a specific mart terminal, A is the number of a plurality of announcements, and p _n is is the number of mandatory announcements completed in a specific mart terminal, q _n is the number of optional announcements completed in a specific mart terminal, p is the number of at least one mandatory announcement, and q is the number of at least one optional announcement It is a number.

8 is a diagram for explaining learning of a first artificial neural network according to an embodiment.

According to an embodiment, the first artificial neural network may be an algorithm that outputs a product inventory update cycle suitable for the product purchase pattern after receiving an analysis result of the product purchase pattern.

The learning device in which the first artificial neural network is learned may be the same device as the device 200 that analyzes the stock update cycle of a product suitable for the purchase pattern of the product using the learned first artificial neural network, or may be a separate device. there is. Hereinafter, a process of learning the first artificial neural network will be described.

First, in step S801, the learning device may generate an input based on an analysis result of a purchase pattern of a product.

Specifically, the learning device may perform a process of pre-processing the analysis result of the purchasing pattern of the product. The analysis result of the embedding pattern of the preprocessed product may be used as an input to the first artificial neural network, or an input may be generated through a normal process of removing unnecessary information.

In step S802, the learning device may apply the input to the first artificial neural network. The first artificial neural network may be an artificial neural network trained according to reinforcement learning. The first artificial neural network may be a Q-Network, a Depp Q-Network (DQN), or a relational network (RL) structure suitable for outputting abstract inference through reinforcement learning.

The first artificial neural network learned through reinforcement learning may be updated and optimized by reflecting evaluations on various rewards. For example, in the first compensation, if the product purchase pattern is purchased at intervals greater than a certain period, the longer the inventory update cycle is selected, the higher the compensation value may be, and in the second compensation, the product purchase pattern is purchased at less than a certain interval. In this case, the shorter the inventory update cycle is selected, the higher the compensation value may be.

In step S803, the learning device may obtain an output from the first artificial neural network. The output of the first artificial neural network may include information about a stock update cycle suitable for a purchase pattern of a product. At this time, the first artificial neural network may select a stock update cycle suitable for the purchase pattern of the product through analysis of the purchase pattern of the product, and output information on the selected stock update cycle.

In step S804, the learning device may evaluate the output of the first artificial neural network and provide a reward. At this time, the evaluation of the output may be divided into first compensation, second compensation, and the like.

Specifically, when the purchase pattern of the product is purchased at intervals greater than or equal to a certain interval, the longer the inventory update cycle is selected, the more first rewards are awarded, and when the purchase pattern of the product is purchased at intervals less than a certain interval, the inventory update cycle is longer. The shorter the update cycle is selected, the more second rewards can be awarded.

In step S805, the learning device may update the first artificial neural network based on the evaluation.

Specifically, the learning device is in a specific state so that the expectation of the sum of rewards is maximized in an environment in which the first artificial neural network selects and extracts the inventory update cycle through the purchase pattern of the product. The first artificial neural network may be updated through a process of optimizing a policy for determining actions to be taken in states.

For example, through the analysis result of the purchasing pattern of the first product, if the inventory update cycle of the first product is set to the first cycle, the learning device can determine whether there is a problem with the analysis result of the first cycle, , If there is no problem with the analysis result of the first cycle, first learning data including information indicating that there is no problem with the analysis of the inventory update cycle of the product is generated, and the first training data is transmitted to the first artificial neural network. By applying, when an embedding pattern similar to the embedding pattern of the first product is input, the first artificial neural network may be updated through a process of learning the first artificial neural network to select a period similar to the first period.

Meanwhile, the process of optimizing the policy may be performed through a process of estimating the maximum value of the expected value of the sum of rewards or the maximum value of the Q-function, or estimating the minimum value of the loss function of the Q-function. Estimation of the minimum value of the loss function may be performed through stochastic gradient descent (SGD) or the like. The process of optimizing the policy is not limited thereto, and various optimization algorithms used in reinforcement learning may be used.

The learning device may gradually update the first artificial neural network by repeating the learning process of the first artificial neural network as described above. Through this, the learning device may learn the first artificial neural network that outputs a stock update cycle suitable for the purchase pattern of the product through the analysis result of the purchase pattern of the product.

That is, when extracting the inventory update cycle through the analysis result of the purchase pattern of the product, the learning device adjusts the analysis criteria by reflecting reinforcement learning through the first compensation and the second compensation, thereby generating the first artificial neural network. can be learned

9 is a diagram for explaining learning of a second artificial neural network according to an embodiment.

According to an embodiment, the second artificial neural network receives as input a matching result of matching the inventory amount, the remaining period until the expiration date, and the daily average sales volume, and the inventory amount may remain without being sold until the expiration date has elapsed. It can be an algorithm that outputs probabilities.

The learning device in which the second artificial neural network is learned may be the same device as the device 200 that analyzes the probability that the stock quantity may remain unsold until the expiration date has elapsed using the learned second artificial neural network, , may be a separate device. Hereinafter, a process of learning the second artificial neural network will be described.

First, in step S901, the learning device may generate an input based on a matching result obtained by matching the inventory amount, the remaining period until the expiration date, and the daily average sales volume.

Specifically, the learning device may perform a process of pre-processing a matching result obtained by matching the inventory amount, the remaining period until the expiration date, and the daily average sales volume. The preprocessed matching result may be used as an input to the second artificial neural network, or an input may be generated through a normal process of removing unnecessary information.

In step S902, the learning device may apply the input to the second artificial neural network. The second artificial neural network may be an artificial neural network trained according to reinforcement learning. The second artificial neural network may be a Q-Network, a Depp Q-Network (DQN), or a relational network (RL) structure suitable for outputting abstract inference through reinforcement learning.

The second artificial neural network learned through reinforcement learning may be updated and optimized by reflecting evaluations on various rewards. For example, if the third compensation is analyzed as having a higher probability that the inventory remains unsold until the expiry date as the amount of inventory increases, the compensation value may increase, and the fourth reward may indicate the remaining amount remaining until the expiration date. If it is analyzed that the shorter the period is, the higher the probability that the stock amount can remain without being sold until the expiration date passes, the compensation value can be high, and the fifth reward is that the smaller the average daily sales volume, the higher the probability of remaining until the expiration date The compensation value can be increased if the inventory is analyzed as having a high probability of remaining unsold.

In step S903, the learning device may obtain an output from the second artificial neural network. The output of the second artificial neural network may be information about the probability that the stock quantity remains unsold until the expiration date passes. At this time, the second artificial neural network analyzes the probability that the inventory amount remains unsold until the expiration date, taking into account the inventory amount, the remaining period until the expiration date, and the daily average sales volume, and determines whether the expiration date has elapsed. Information on the probability that the stock will remain unsold until

In step S904, the learning device may evaluate the output of the second artificial neural network and provide a reward. At this time, the evaluation of the output may be divided into third compensation, fourth compensation, fifth compensation, and the like.

Specifically, if the learning device analyzes that the greater the amount of inventory, the higher the probability that the amount of inventory can remain without being sold until the expiration date passes, the third reward is awarded more, and the shorter the remaining period remaining until the expiration date, the more distribution If it is analyzed that there is a high probability that the stock will remain unsold until the expiration date, the fourth reward will be awarded more, and the smaller the average daily sales volume, the more likely the stock will remain unsold until the expiration date. If analyzed as having a high probability, a lot of fifth rewards can be awarded.

In step S905, the learning device may update the second artificial neural network based on the evaluation.

Specifically, the learning device generates an expected value of the sum of reward values in an environment in which the second artificial neural network analyzes the probability that the stock quantity remains unsold until the expiration date elapses In order to maximize this, the second artificial neural network may be updated through a process of optimizing a policy for determining actions to be taken in specific states.

For example, the learning device determines a probability that the first inventory amount remains unsold until the first period elapses through a first matching result obtained by matching the first inventory amount, the first period, and the first sales amount. If the first occurrence rate is predicted, it is possible to check whether or not there is a problem with the analysis result of the first incidence rate, and if there is no problem with the analysis result of the first incidence rate, the inventory remains unsold until the expiration date has elapsed. Second training data including information indicating that there is no problem in the analysis of possible probability is generated, and the second training data is applied to the second artificial neural network to determine the first inventory, the first period, and the first sales volume. When a matching result similar to the matched first matching result is input, the second artificial neural network may be updated through a process of training the second artificial neural network to select a probability similar to the first occurrence rate.

Meanwhile, the process of optimizing the policy may be performed through a process of estimating the maximum value of the expected value of the sum of rewards or the maximum value of the Q-function, or estimating the minimum value of the loss function of the Q-function. Estimation of the minimum value of the loss function may be performed through stochastic gradient descent (SGD) or the like. The process of optimizing the policy is not limited thereto, and various optimization algorithms used in reinforcement learning may be used.

The learning device may gradually update the second artificial neural network by repeating the above-described learning process of the second artificial neural network. Through this, the learning device will train a second artificial neural network that outputs the probability that the inventory amount will remain unsold until the expiration date, taking into account the inventory amount, the remaining period until the expiration date, and the average daily sales volume. can

That is, when the learning device analyzes the probability that the inventory amount may remain unsold until the expiration date passes through the matching result of matching the inventory amount, the remaining period until the expiration date, and the daily average sales volume, the third The second artificial neural network may be trained by adjusting the analysis criterion by reflecting reinforcement learning through the reward, the fourth reward, the fifth reward, and the like.

10 is an exemplary diagram of a configuration of a device according to an embodiment.

Device 200 according to an embodiment includes a processor 210 and a memory 220 . The processor 210 may include at least one device described above with reference to FIGS. 1 to 9 or may perform at least one method described above with reference to FIGS. 1 to 9 . A person or organization using the device 200 may provide services related to some or all of the methods described above with reference to FIGS. 1 to 9 .

The memory 220 may store information related to the methods described above or may store a program in which the methods described below are implemented. Memory 220 may be volatile memory or non-volatile memory.

The processor 210 may execute a program and control the device 200 . Program codes executed by the processor 210 may be stored in the memory 220 . The device 200 may be connected to an external device (eg, a personal computer or network) through an input/output device (not shown) and exchange data through wired/wireless communication.

Apparatus 200 may be used to train an artificial neural network or to use a trained artificial neural network. The memory 220 may include a learning or learned artificial neural network. The processor 210 may learn or execute an artificial neural network algorithm stored in the memory 220 . The device 200 for learning the artificial neural network and the device 200 using the learned artificial neural network may be the same or may be separate.

The embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. You can command the device. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

Claims (3)

In the distribution management method of inventory for artificial intelligence-based mart operation, performed by the device,
Analyzing a purchase pattern of purchasing a first product at the first mart based on purchase details of the first mart;
setting a first period, which is a stock update period of the first product, based on an output of the first artificial neural network by applying an analysis result of the purchase pattern of the first product to a first artificial neural network;
classifying the first products stored in the first mart by expiration date for each first cycle, and classifying products having the shortest remaining period until expiration date among the first products into a first group;
identifying a stock amount of the first product classified into the first group as a first stock amount, and determining a remaining period until an expiration date of the first product classified into the first group as a first period; and
When it is determined that the first period is shorter than a preset first reference period, sending a notification message informing that the expiration date for the first stock amount of the first product in stock is imminent to a first mart terminal. do,
Transmitting a notification message informing that the expiration date for the first inventory amount is imminent to the first mart terminal,
determining the daily average sales volume of the first product as a first sales volume based on the sales details of the first mart;
generating a first matching result by matching the first inventory amount, the first period, and the first sales volume;
By applying the first matching result to the second artificial neural network, based on the output of the second artificial neural network, the probability that the first inventory amount remains unsold until the first period elapses is determined by first predicting by incidence;
setting a distribution management state for the first stock amount to a warning state when it is confirmed that the first occurrence rate is equal to or greater than a preset first reference value;
setting a distribution management state for the first inventory amount to a caution state when it is determined that the first occurrence rate is less than the first reference value and equal to or greater than a preset second reference value;
setting a distribution management state for a first stock amount to a normal state when it is confirmed that the first occurrence rate is less than the second reference value;
Transmitting a notification message recommending a discount for the first product to the first mart terminal when the distribution management state for the first stock amount is set to a warning state;
transmitting a notification message recommending a promotion for the first product to the first mart terminal when the distribution management state for the first stock amount is set to a caution state; and
When the distribution management state for the first inventory is set to a normal state, transmitting a notification message recommending purchase of the first product to the first mart terminal,
Inventory distribution management method for artificial intelligence-based mart operation. delete According to claim 1,
Transmitting a notification message recommending a discount for the first product to the first mart terminal,
setting a first ratio to 99% when it is confirmed that the first stock amount is less than a preset first reference amount;
setting the first ratio to a value lower than 99% as the first inventory amount increases when it is determined that the first stock amount is equal to or greater than the first reference amount;
setting a second ratio to 99% when it is determined that the first period is longer than a preset second reference period;
if it is determined that the first period is not longer than the second reference period, setting the second ratio to a value lower than 99% as the first period is shorter;
calculating a second price by multiplying the first price, the first ratio, and the second ratio when the price of the first product is set as the first price; and
Changing the price of the first product from the first price to the second price and transmitting a notification message recommending a discount for the first product to the first mart terminal,
Inventory distribution management method for artificial intelligence-based mart operation.

Images