KR20210045346A - Method and apparatus for forecasting shipments of warehouse using artificial intelligence model - Google Patents

Abstract

Provided is a method for predicting shipment volume for each item in a warehouse by using an artificial intelligence model. According to one disclosure, the method, which is performed by a computer, comprises the following steps of: obtaining information on the delivery history of a first item stored in a warehouse; obtaining information on a plurality of external factors related to the time of shipment of the first item; generating a relationship graph for a plurality of external factors having an effect on the shipment of the first item by calculating a weight for each of the external factors occurring at the time of shipment of the first item by using a first artificial intelligence model; and determining the predicted shipment amount of the first item in consideration of a plurality of external factors that change according to a time-series flow by using the relationship graph.

Description

A method and device for predicting shipments by item in a distribution warehouse using an artificial intelligence model {METHOD AND APPARATUS FOR FORECASTING SHIPMENTS OF WAREHOUSE USING ARTIFICIAL INTELLIGENCE MODEL}

The present invention relates to a method and apparatus for predicting the amount of shipment for each item in a distribution warehouse in consideration of external factors using an artificial intelligence model.

Artificial intelligence technology includes machine learning (machine learning) technology, and among machine learning technologies, it includes deep learning technology, which is particularly widely used to analyze images.

Deep learning is a set of machine learning algorithms that attempts high-level abstractions (summarizing key contents or functions in large amounts of data or complex data) through a combination of several nonlinear transducers. Is defined. Deep learning can be seen as a branch of machine learning that teaches computers how people think in a big frame.

When there is any data, it is represented in a form that can be understood by a computer (for example, in the case of an image, pixel information is expressed as a column vector, etc.), and a lot of research (how to do better expression) to apply it to learning. How to create techniques and how to create models to learn them) is in progress. As a result of these efforts, various deep learning techniques have been developed. Examples of deep learning techniques include Deep Neural Networks (DNN), Convolutional deep Neural Networks (CNN), Reccurent Neural Networks (RNN), and Deep Belief Networks (DBN). have.

Deep Neural Networks (DNNs) are artificial neural networks (ANNs) composed of a plurality of hidden layers between an input layer and an output layer.

These artificial intelligence technologies are rapidly being applied to various fields, but despite their necessity, they are still not properly applied to industrial sites in many cases.

According to the industrial development, the production of items in various industries has increased, and accordingly, distribution warehouses capable of storing items scheduled for distribution have increased. However, if the quantity of items stored in the distribution warehouse cannot be managed, the distribution warehouse cannot be used efficiently, resulting in damage. Accordingly, there has been an increase in the need for a method that can effectively use the distribution warehouse by predicting the delivery amount of items stored in the distribution warehouse.

The problem to be solved by the present invention is to provide a method of predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

According to the first embodiment of the present invention for solving the above-described problem, the step of acquiring information on the delivery history of the first item stored in the distribution warehouse, and obtaining information on a plurality of external factors related to the delivery time of the first item A step of calculating a weight for each of a plurality of external factors that occurred at the time of shipment of the first item using the first artificial intelligence model, and generating a relationship graph for the external factors affecting the shipment of the first item; and Providing a method for predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model, including the step of determining the predicted shipment amount of the first item in consideration of a plurality of external factors that change according to a time series flow using a relationship graph. I can.

In order to solve the above-described problem, an apparatus for predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to an aspect of the present invention includes a memory for storing one or more instructions and a processor for executing the one or more instructions stored in the memory. And the processor executes the one or more instructions to perform the method according to the disclosed embodiment.

In order to solve the above-described problems, the program for predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model according to an aspect of the present invention is combined with a computer, which is hardware, and read from a computer to perform the method according to the disclosed embodiment. It is stored on a recording medium where possible.

Other specific details of the present invention are included in the detailed description and drawings.

According to the disclosed embodiment, it is possible to provide information necessary for the operation and management of the distribution warehouse by predicting the amount of delivery of a distribution warehouse using an artificial intelligence model, and by predicting sales of the distribution warehouse.

According to one disclosure, according to the present invention, by using a graph of a relationship between a plurality of external factors acquired through an artificial intelligence model and a shipment amount of an item, it is possible to predict the shipment amount of a future item in consideration of external factors. In addition, it is possible to obtain more detailed and accurate information on the predicted shipment quantity by compensating for the difference between the actual shipment quantity and the predicted shipment quantity.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

FIG. 1 is a diagram illustrating a system for predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to an exemplary embodiment.
FIG. 2 is a diagram schematically illustrating a method of predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to an exemplary embodiment.
3 is a flowchart illustrating a method of predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to the start of the day.
FIG. 4 is a diagram for explaining a feature of predicting an item's delivery amount based on a graph of a relationship between an item and an external factor according to one start.
5 is a diagram showing the configuration of a device for predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to the start of the day.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used herein, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements. Throughout the specification, the same reference numerals refer to the same elements, and "and/or" includes each and all combinations of one or more of the mentioned elements. Although "first", "second", and the like are used to describe various elements, it goes without saying that these elements are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical idea of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used with meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically.

The term "unit" or "module" used in the specification refers to a hardware component such as software, FPGA or ASIC, and the "unit" or "module" performs certain roles. However, "unit" or "module" is not meant to be limited to software or hardware. The “unit” or “module” may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example, "sub" or "module" refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays and variables. Components and functions provided within "sub" or "modules" may be combined into a smaller number of components and "sub" or "modules" or into additional components and "sub" or "modules". Can be further separated.

In the present specification, a computer means all kinds of hardware devices including at least one processor, and may be understood as encompassing a software configuration operating in the corresponding hardware device according to embodiments. For example, the computer may be understood as including all of a smartphone, a tablet PC, a desktop, a laptop, and a user client and application running on each device, and is not limited thereto.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Each of the steps described herein is described as being performed by a computer, but the subject of each step is not limited thereto, and at least some of the steps may be performed by different devices according to embodiments.

Hereinafter, the device 100 for predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model will be abbreviated as the device 100 for predicting the amount of item shipped.

1 is a diagram illustrating a system for predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to an exemplary embodiment.

Referring to FIG. 1, a system and components thereof are disclosed for explaining a method of predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to the disclosed embodiment.

The system shown in FIG. 1 includes a distribution warehouse 10, an item delivery quantity prediction apparatus 100, and a user terminal 200.

In one embodiment, the type of distribution warehouse indicated by the distribution warehouse 10 is not limited, and the distribution warehouse 10 collects information such as the type of items stored in the distribution warehouse 10, inventory, stocking and delivery quantity, etc. It may include a management terminal to manage.

The item delivery quantity prediction device 100 collects information on the distribution warehouse 10 from the management terminal of the distribution warehouse 10, and generates information necessary for management, operation, and management of the distribution warehouse 10 based on this. have. The item shipment quantity prediction device 100 may include a server, a mobile terminal, a cloud, an electronic device, etc., and includes all types of devices capable of executing the method of predicting the shipment quantity for each item in a distribution warehouse using the artificial intelligence of the present application model. can do.

The information generated by the item shipment amount prediction apparatus 100 may be transmitted to the user terminal 200. For example, the user terminal 200 may mean a terminal used by a manager or manager of the distribution warehouse 10, but is not limited thereto.

In an embodiment, the item shipment quantity prediction apparatus 100 may predict the shipment quantity of all items included in the distribution warehouse 10. By the start of the day, the item delivery quantity prediction apparatus 100 calculates the estimated delivery quantity of the item based on the correlation between the items included in the distribution warehouse 10 and the delivery of the item due to factors outside the distribution warehouse 10. You can decide.

In an embodiment, the item delivery quantity prediction apparatus 100 may perform a delivery quantity prediction based on delivery history information of a plurality of items.

In an embodiment, the item shipment quantity prediction apparatus 100 may collect information on a daily shipment quantity of a specific item stored in the distribution warehouse 10.

In an embodiment, the item shipment quantity prediction apparatus 100 may predict a shipment quantity at a future point of time based on information on the daily shipment quantity of a corresponding item. For example, the item delivery quantity prediction apparatus 100 predicts the delivery quantity at a future date based on the change in delivery quantity of the corresponding item, or a specific day of the week, a specific date for each month, a specific time for each quarter or semi-annual, and a specific period per year. It collects and manages information on the quantity of shipments such as the time of day or time, and it is also possible to predict the quantity of shipments at a time in the future based on information on the quantity of shipments in the past corresponding to the time in the future.

Specifically, if you want to predict the shipment volume for a specific date, shipments from one or more past times that can be referenced, such as shipments for the date one year ago, shipments for the day one month ago, shipments for the day one week ago Shipments can be predicted based on the change trend.

A method of predicting a shipment amount at a future point based on a shipment quantity history may be performed by learning an artificial intelligence model using training data acquired based on the shipment quantity history. For example, a machine learning technique may be used for learning an artificial intelligence model, but is not limited thereto.

In an embodiment, the item shipment quantity prediction apparatus 100 may perform a shipment quantity prediction based on an external factor. In one embodiment, the item shipment quantity prediction apparatus 100 collects external information such as the day of the week, the weather, the season at a future point in time, and whether or not a promotion for the item is progressed, and analyzes the correlation between each external information and the shipment quantity. And learn.

The item shipment quantity prediction apparatus 100 may predict the shipment quantity of a specific item based on external factors at a point in time in the future based on the learning result.

The item delivery quantity prediction apparatus 100 may perform an operation for inventory and logistics management, such as performing a preemptive stocking request based on the predicted delivery quantity information.

In addition, the item shipment quantity prediction apparatus 100 may obtain information for updating the learned model by storing the predicted data in a database, obtaining actual shipment quantity information at a corresponding point in time, and comparing it with the predicted data.

According to an embodiment, the item shipment amount prediction apparatus 100 may also predict the vacancy rate of the distribution warehouse 10.

In an embodiment, the item shipment quantity prediction apparatus 100 may record the daily vacancy rate of the distribution warehouse 10 and use this to predict the vacancy rate at a future point in time.

For example, the item shipment quantity prediction apparatus 100 may predict the vacancy rate at a future point in time using the change trend of the daily vacancy rate and the vacancy rate at a past point in time, similar to the above-described history-based shipment quantity prediction method.

In addition, the item delivery quantity prediction device 100 collects information on one or more other warehouses 1, and the location, construction outline, and the vacancy rate and distribution of other warehouses having a major storage item are similar to the warehouse 10. The vacancy rate of the warehouse 10 can be compared.

When there is a difference in the vacancy rate between the distribution warehouse 10 and other warehouses (for example, when the vacancy rate of other warehouses is low), the item shipment quantity prediction device 100 is used for the distribution warehouse 10 and other warehouses. The difference between the two can be analyzed, and the cause of vacancy in the distribution warehouse 10 can be analyzed based on this.

For example, the cause of vacancy in the distribution warehouse 10 may include rental fees, rental periods, facilities, etc., but is not limited thereto.

In addition, the item shipment quantity prediction apparatus 100 may predict sales of the distribution warehouse 10 based on the vacancy rate of the distribution warehouse 10. In this case, the item delivery quantity prediction apparatus 100 also acquires information on the distribution warehouse with the highest sales among other distribution warehouses having a location and land area similar to that of the distribution warehouse 10, and determines the difference with the distribution warehouse 10. The analysis may be performed and the corresponding information may be provided to the user terminal 200.

In addition, information provided to the user terminal 200 may be used as information for remodeling the distribution warehouse 10.

FIG. 2 is a diagram schematically illustrating a method of predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to an exemplary embodiment.

By the start of the day, the item delivery quantity prediction apparatus 100 generates predicted delivery quantity information for each of the plurality of items by using the delivery history information for each of the plurality of items stored in the distribution warehouse and external factors generated outside the distribution warehouse. can do.

According to an initiation, the item shipment quantity prediction apparatus 100 may generate predicted shipment quantity information of a specific item at a specific time point by using an artificial intelligence model.

The artificial intelligence model is trained to predict the predicted shipment amount at a future point of time by using information acquired inside the distribution warehouse and information acquired outside the distribution warehouse. In the artificial intelligence model, learning conditions may vary according to conditions predetermined by the user, such as a specific period, a specific event, a specific external factor, or a specific item.

According to the start of the day, the item shipment quantity prediction apparatus 100 uses the learned artificial intelligence model, from the shipment history information of the first item and external factors, the predicted shipment quantity of the first item at a predetermined time or period of the first item. Can generate information.

In addition, according to the start of the day, the item delivery quantity prediction apparatus 100 predicts the occurrence time of each external factor and the delivery information of the items by using the delivery history information of the first item and the second item and a plurality of external factors. It is possible to predict the likelihood that item 1 and item 2 will affect each shipment.

3 is a flowchart illustrating a method of predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to the start of the day.

By the start of the day, in block 301, the item delivery amount prediction apparatus 100 may obtain the delivery history information of the first item stored in the distribution warehouse. By commencement, the delivery history information may include history information from the time of stocking of the item to the time of delivery of the item. In more detail, the delivery history information may include information on the type of item, the quantity received, the property of the item, the property of the buyer group, the stocking period, the delivery time, the sale method, the stock quantity, the promotion status, etc., and the stocking time of the item. It can include all the history information that has occurred.

According to an initiation, in block 302, the apparatus 100 for predicting an item shipment amount may obtain information on a plurality of external factors related to a shipment time of the first item.

By initiation, the external factor is an attribute originating from outside the distribution warehouse, and may include all external factors occurring at the time the item is shipped. For example, external factors may include seasonal information, weather information, temperature information, day of the week information, event information, customer information, sales method information, and attribute information of the warehouse. It will be apparent to those skilled in the art that information may be included.

By the start of the day, the item shipment quantity prediction apparatus 100 includes seasonal information, weather information, temperature information, day of the week information, event information, purchase customer information, sales method information, and attribute information of the distribution warehouse at the time of shipment of the first item. It is possible to obtain information on external factors that do. The item shipment prediction device 100 may receive external factors from the meteorological service server, media server, shopping mall server, advertisement server, user terminal, SNS server, etc., and temperature and humidity around the distribution warehouse from sensor devices installed around the distribution warehouse. It is also possible to obtain external factors such as the concentration of fine dust, air pressure, and whether or not there is a disaster. A method of obtaining an external factor by the item shipment quantity prediction apparatus 100 is not limited.

By the start of the day, in block 303, the item shipment amount prediction apparatus 100 calculates a weight for each of a plurality of external factors occurring at the time of shipment of the first item using the first artificial intelligence model to influence the shipment of the first item. You can create a graph of the relationship for external factors that affect.

According to an initiation, the apparatus 100 for predicting an item shipment amount may assign a weight to each of a plurality of external factors based on the frequency of the external factors occurring at each shipment time of the first item. For example, in the case where the first item is shipped 5 times, the weight of the external factor having the highest overlapping frequency among the external factors occurring at each delivery time is the highest, and the lower the frequency of occurrence, the lower the weight.

According to an initiation, the apparatus 100 for predicting an item delivery amount may determine a predicted delivery amount of the first item by analyzing a relationship between the delivery time of the first item and a plurality of external factors, based on the assigned weight. The item shipment amount prediction apparatus 100 may determine the predicted shipment amount at a specific time in the future by the start of the day, and may determine the predicted shipment amount in units of a predetermined period.

According to an initiation, in block 304, the apparatus 100 for predicting an item shipment amount may determine a predicted shipment amount of the first item in consideration of a plurality of external factors that change according to a time series flow using a relationship graph.

According to an initiation, the item shipment amount prediction apparatus 100 may determine a stocking amount and a stocking time of the first item based on the predicted shipment amount of the first item.

In addition, the item delivery quantity prediction apparatus 100 may modify the stocking quantity and the stocking time of the first item in response to the occurrence of an irregular flow of at least one external factor. For example, based on the three-year's worth of item shipment history, the estimated shipment volume of air conditioners from June to August was determined based on the increase in the sales volume of air conditioners when the heat began. If it is expected to occur, the occurrence of a typhoon can be determined as an irregular flow. Accordingly, the predicted shipment amount of the air conditioner item can be adjusted to be lower than the existing predicted shipment amount.

By the start of the day, the item delivery quantity prediction apparatus 100 generates a first event from a combination of a plurality of external factors acquired at the first delivery time when the delivery quantity of the first item is detected above the threshold value using the second artificial intelligence model. can do. Here, the first event may represent an event that can occur naturally according to the passage of time. For example, when sales of air conditioners increase as the tropical night continues, the item shipment quantity predicting device 100 may determine the temperature and the temperature according to the tropical night. A humidity increase event may be generated, and when the event is generated, it may be predicted that the predicted shipment amount of the air conditioner may increase. Here, the threshold value is an index for indicating a case in which the demand for a specific item increases rapidly, and may be determined based on the stock quantity according to the stock quantity of the item, reorder information, etc.

For example, the item shipment quantity prediction apparatus 100 may provide a plurality of information on weather information, Christmas season information, fashion of a fur hat in a movie, etc., at a time when the shipment quantity of a fur hat item is detected above a threshold value. It is possible to create a “winter-hair hat event” from a combination of external factors.

The item delivery amount prediction apparatus 100 may predict the occurrence time of the first event using the second artificial intelligence model, and adjust the stocking amount of the first item at the predicted time. That is, the item shipment quantity prediction apparatus 100 repeats the process of determining the predicted shipment quantity and the predicted shipment time of the first item using the data acquired during a predetermined period, and re-determining the predicted shipment quantity using the newly acquired data. By doing so, it is possible to obtain more reliable predicted shipments. For example, based on a plurality of factors for winter, cold, Christmas gifts, etc. included in the “winter-hairy hat event”, the amount of woolen hats worn in December may be increased.

In more detail, the item shipment amount prediction apparatus 100 may calculate a difference value between the predicted shipment amount of the first item and the actual shipment amount by using the shipment history information of the first item acquired every predetermined period.

In this case, the item shipment amount prediction apparatus 100 may re-determine the weights of a plurality of external factors that have influenced the difference value. In addition, the item shipment amount prediction apparatus 100 may modify the predicted shipment amount of the first item based on the weights of a plurality of re-determined external factors.

For example, when it is determined that the quantity of fur hats shipped in winter will increase, and the quantity of wearing of 100 fur hats is determined, but only 40 fur hats are shipped in the winter of that year due to an abnormal high temperature phenomenon, the item shipment quantity prediction device (100) may reset the weight according to the “winter temperature”, and re-determine the amount of wearing of the fur hat based on winter temperature prediction information acquired in the following winter. According to an embodiment, a weight re-determination for one external factor has been described, but it is obvious to those skilled in the art that it can be simultaneously applied to a plurality of external factors.

FIG. 4 is a diagram for explaining a feature of predicting an item's shipment amount by using a graph of a relationship between an item and an external factor according to an initial date.

According to an initiation, the apparatus 100 for predicting an item shipment amount may generate a plurality of nodes corresponding to a plurality of external factors occurring at the time of shipment of the first item in order to generate a relationship graph. In this case, the first item may also be displayed as a node. Like the relationship graph shown in FIG. 4, the first item (ex. travel bag) may be displayed as one node in the relationship graph. In addition, a plurality of external factors acquired at each shipment point of the first item may be displayed as nodes of the relationship graph. External factors may include external factors occurring in distribution warehouses such as days of the week, typhoon weather, vacation, summer, hot and humid weather, and overseas travel.

According to an initiation, the apparatus 100 for predicting an item shipment amount may generate at least one line connecting each of the plurality of external factors in order to represent the relationship between each of the plurality of external factors. The plurality of external factors are all information on external attributes generated at the time of shipment of the first item, and the plurality of external factors generated at the time of shipment of the first item may be directly, indirectly and organically connected.

According to an initiation, the apparatus 100 for predicting an item shipment amount may adjust the thickness of at least one line to indicate the strength of a relationship between a plurality of external factors simultaneously occurring at the time of shipment of the first item. For example, in the case of external factors occurring at the same time when the first item is shipped, it may be determined that the strength of the correlation between the two external factors is high, and the thickness of a line connecting the two external factors may be adjusted accordingly. For example, if the frequency of overseas travel, summer, and vacation external factors affecting the time of shipment of a suitcase was high, increase the thickness of the line connecting the suitcase-overseas travel-summer-vacation, respectively, The strength of the relationship between nodes can be increased.

In an embodiment, a relationship between a specific item and an external factor may be displayed and stored using various types of lines as well as thickness. For example, if there is insufficient information to determine the relationship between a specific item and an external factor (for example, when the number of collected cases is less than a preset number), draw a line connecting the item and the node corresponding to an external factor. It can be marked with a broken line.

As another example, when a relationship between a specific item and an external factor appears differently under a specific condition, the shape of a line connecting the corresponding nodes may be differently set. For example, if the frequency of external factors such as overseas travel, summer, and vacation is high among external factors that affect the time of shipment of a suitcase, but the frequency of the external factor is low only when the amount of precipitation is more than a preset value, the suitcase- The shape of the line connecting overseas travel-summer-vacation can be made into a wave shape. In addition, a wave-shaped filter corresponding to a day when the precipitation is greater than or equal to a preset value may be stored. On the other hand, an item irrelevant to precipitation and a node corresponding to an external factor may be connected in a straight line.

In an embodiment, the apparatus 100 for predicting an item shipment amount may search for an item for which a correlation is to be determined and a node corresponding to one or more external factors in a relationship graph, and obtain information on a thickness of a line between the searched nodes. The item shipment amount prediction apparatus 100 may determine a correlation between the item and an external factor based on the acquired thickness information.

In addition, the item shipment amount prediction apparatus 100 may obtain information on a characteristic (a broken line, etc.) of a line between nodes retrieved from a relation graph, and obtain information on a correlation corresponding to the corresponding characteristic.

In addition, the item shipment amount prediction apparatus 100 may apply a pre-stored wave-type filter to all lines included in the graph when attempting to obtain a correlation between an item and an external factor on a day when the amount of precipitation is greater than or equal to a preset value. have. Through the application of the filter, a line of a portion overlapping a previously stored wave shape may be deleted, a color may be blurred, or a thickness may be reduced. For example, in the case of a straight line, only a portion overlapping with the wave shape of the filter may be deleted to form a broken line, and in the case of a wave type line, the entire line may be deleted by overlapping with the wave shape of the filter.

The item shipment amount prediction apparatus 100 may determine that the correlation is high when there are few deleted portions, and that the correlation is high as the thickness of the undelete portion increases, based on the state of the lines of the graph to which the filter is applied. .

By the start of the day, the item delivery quantity prediction apparatus 100 predicts the estimated delivery quantity of an item from item delivery history information and a plurality of external factors through an artificial intelligence model, determines the quantity and time of stocking of the item, or stores the item. It is possible to re-determine the item arrangement information in the distribution warehouse. In addition, by predicting the sales amount of the distribution warehouse according to the prediction of the item's warehousing and delivery, it is possible to determine monitoring information that can efficiently utilize the distribution warehouse.

For example, the item shipment quantity prediction apparatus 100 may determine that summer, vacation, high temperature and high humidity, and overseas travel as summer vacation events, and predict that the shipment amount of suitcases will increase in early July. In addition, the item shipment quantity prediction apparatus 100 may determine how much to wear the travel bag at a specific time before the time when the item shipment quantity is predicted to increase.

In addition, the item shipment quantity prediction apparatus 100 may determine a predicted shipment quantity of an item by analyzing a plurality of factors at a time when the item shipment quantity is high. In more detail, the apparatus 100 for predicting an item shipment amount may determine a first period as a period during which the shipment amount of the first item is the highest. Here, the first period may be determined by the user, or may be determined as a time when the quantity of the first item is measured higher than a predetermined threshold.

The item delivery amount prediction apparatus 100 may determine at least one first external factor calculated with a weight equal to or greater than a threshold value during the first period using the relationship graph. Here, the weight greater than or equal to the threshold means that an external factor has occurred more than a predetermined number of times during the first cycle, and the external factor determined by the weight greater than or equal to the threshold can be determined to have a great influence on the high quantity of items shipped during the first cycle. . Here, the first external factor may include at least one external factor.

In addition, the item shipment amount prediction apparatus 100 may determine a predicted shipment amount of an item by analyzing a plurality of factors at a time when the item shipment amount is low. In more detail, the item shipment amount prediction apparatus 100 may determine a second period as a period in which the shipment amount of the first item is the lowest. Here, the second period may be determined by the user, or may be determined as a time when the delivery amount of the second item is measured lower than a predetermined threshold value.

The item delivery amount prediction apparatus 100 may determine at least one second external factor calculated with a weight equal to or greater than a threshold value during the second period using the relationship graph. Here, the weight greater than or equal to the threshold means that an external factor has occurred more than a predetermined number of times during the second period, and the external factor determined by the weight greater than or equal to the threshold can be determined to have a great influence on the low quantity of items shipped during the second period. . Here, the second external factor may include at least one external factor.

According to an initiation, the apparatus 100 for predicting an item shipment amount may determine a second item whose shipment amount is increased when at least one determined first external factor occurs. For example, in a situation in which the first external factor occurs and the amount of shipment of the second item is high, it can be seen that the first external factor affects not only the first item, but also the delivery of the second item. For example, when the shipment volume of a suitcase is increased, the external factors having high weights are summer, vacation, and overseas travel, and at the same time, when the shipment volume related to swimming goods increases, the swimsuit may be determined as the second item.

When the occurrence of at least one first external factor is predicted by an initiation, the item delivery quantity prediction apparatus 100 may adjust the stocking quantity of the first item and the second item. Accordingly, according to the present invention, by analyzing the relationship between a plurality of items using an artificial intelligence model, it is possible to determine a predicted shipment amount of a plurality of items according to a change in a plurality of external factors.

In addition, the item delivery amount prediction apparatus 100 may generate first item arrangement information of a distribution warehouse using location information and quantity information of the first item. The item delivery quantity prediction apparatus 100 may acquire arrangement information of at least one item arranged in a distribution warehouse in real time. By the start of the day, the item delivery quantity prediction apparatus 100 may arrange the arrangement of the items with the early predicted delivery time to a position easier to carry than the items with the late delivery time of the prediction.

In addition, the item delivery amount prediction apparatus 100 may modify the first item arrangement information in the distribution warehouse using the predicted delivery amount and the predicted delivery time of the first item. For example, when the delivery timing of the first item is modified, the arrangement of the items may be changed for efficient logistics warehouse operation.

5 is a diagram showing the configuration of a device for predicting a shipment amount for each item of a distribution warehouse using an artificial intelligence model according to the start of the day.

According to one disclosure, the apparatus 100 for predicting an item shipment amount may include a memory 1200 that stores one or more instructions and a processor 1100 that executes one or more instructions stored in the memory 1200. According to one disclosure, the processor 1100 included in the apparatus 100 for predicting an item shipment amount may perform the method of the present application by executing one or more instructions.

However, the item shipment amount prediction apparatus 100 may be implemented by more components than the components illustrated in FIG. 5, or may be implemented by fewer components.

The processor 1100 controls the overall operation of the apparatus 100 for predicting the amount of item shipped. For example, the processor 1100 may generally control other components included in the item shipment amount prediction apparatus 100 by executing programs stored in the memory 1200. In addition, the processor 1100 may perform a function of the apparatus 100 for predicting an item shipment amount by executing programs stored in the memory 1200. The processor 1100 may include at least one processor. The processor 1100 may include a plurality of processors or may include a single processor in an integrated form according to its function and role. In an embodiment, the processor 1100 may include at least one processor configured to provide a notification message by executing at least one program stored in the memory 1200.

The memory 1200 may store a program for processing and controlling the processor 1100, and may store data input to the item shipment quantity prediction apparatus 100 or output from the item shipment quantity prediction apparatus 100.

The memory 1200 is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (for example, SD or XD memory 1200). ), RAM (RAM, Random Access Memory), SRAM (Static Random Access Memory), ROM (ROM, Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), Magnetic It may include at least one type of storage medium among memory, magnetic disk, and optical disk.

Programs stored in the memory 1200 may be classified into a plurality of modules according to their functions. Here, the plurality of modules are software, not hardware, and refer to modules that operate functionally.

The processor 1100 according to an embodiment may include a data learning unit 11001 and a data recognition unit 1102.

At least one of the data learning unit 1101 and the data recognition unit 1102 may be manufactured in the form of at least one hardware chip and mounted on an electronic device. For example, at least one of the data learning unit 1101 and the data recognition unit 1102 may be manufactured in the form of a dedicated hardware chip for artificial intelligence (AI), or an existing general-purpose processor (eg, a CPU Alternatively, it may be manufactured as a part of an application processor) or a graphics dedicated processor (eg, a GPU) and mounted on the aforementioned various electronic devices.

In this case, the data learning unit 1101 and the data recognition unit 1102 may be mounted on one electronic device, or may be mounted on separate electronic devices, respectively. For example, one of the data learning unit 1101 and the data recognition unit 1102 may be included in the electronic device, and the other may be included in the server. In addition, the data learning unit 1101 and the data recognition unit 1102 may provide model information built by the data learning unit 1101 to the data recognition unit 1102 through wired or wireless communication, or the data recognition unit ( Data input to 1102 may be provided to the data learning unit 1101 as additional learning data.

Meanwhile, at least one of the data learning unit 1101 and the data recognition unit 1102 may be implemented as a software module. When at least one of the data learning unit 1101 and the data recognition unit 1102 is implemented as a software module (or a program module including an instruction), the software module is computer-readable and non-transitory. It may be stored in a non-transitory computer readable media. Also, in this case, at least one software module may be provided by an operating system (OS) or may be provided by a predetermined application. Alternatively, some of the at least one software module may be provided by an operating system (OS), and the remaining part may be provided by a predetermined application.

The above-described method for predicting the amount of shipment for each item of the distribution warehouse may be performed based on an artificial intelligence model learned through data including information on the existing history, one or more external factors, and information on the amount of shipment for each item accordingly. The learning method of the artificial intelligence model and the types of training data used therein are not limited.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. Software modules include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

Components of the present invention may be implemented as a program (or application) and stored in a medium in order to be combined with a computer that is hardware to be executed. Components of the present invention may be implemented as software programming or software elements, and similarly, embodiments include various algorithms implemented with a combination of data structures, processes, routines or other programming elements, including C, C++ , Java, assembler, or the like may be implemented in a programming or scripting language. Functional aspects can be implemented with an algorithm running on one or more processors.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. You will be able to understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

Claims (10)

In the method performed by a computer,
Acquiring delivery history information of the first item stored in the distribution warehouse;
Acquiring information on a plurality of external factors related to the delivery time of the first item;
Generating a relationship graph for a plurality of external factors affecting the delivery of the first item by calculating weights for each of a plurality of external factors occurring at the delivery time of the first item using the first artificial intelligence model; And
And determining the predicted shipment amount of the first item by using the relationship graph. A method of predicting shipment amount for each item of a distribution warehouse using an artificial intelligence model. The method of claim 1,
Obtaining information on external factors including seasonal information, weather information, temperature information, day of the week information, event information, purchase customer information, sales method information, and attribute information of a distribution warehouse of the first item at the time of shipment of the first item;
Assigning a weight to each of the plurality of external factors based on the frequency of the external factors occurring at each delivery time of the first item; And
A distribution warehouse using an artificial intelligence model comprising the step of determining a predicted delivery amount of the first item by analyzing a relationship between the delivery time of the first item and the plurality of external factors based on the assigned weight. How to predict the amount of shipment by item. The method of claim 1,
Generating a first event from a combination of a plurality of external factors acquired at a first delivery time when the delivery amount of the first item is detected to be greater than or equal to a threshold value using a second artificial intelligence model; And
Predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model, including the step of predicting the occurrence time of the first event using the second artificial intelligence model and adjusting the amount of stocking of the first item at the predicted time point Way. The method of claim 1,
Calculating a difference value between a predicted shipment amount of the first item and an actual shipment amount of the first item by using the shipment history information of the first item acquired every predetermined period;
Re-determining weights of a plurality of external factors associated with the difference value; And
And adjusting the predicted shipment amount of the first item based on the re-determined weights of the plurality of external factors. The method of claim 1,
Determining a first period of a period in which the quantity of the first item is shipped is the highest;
Determining at least one first external factor calculated as a weight greater than or equal to a threshold value during the first period using the relationship graph;
Determining a second item for which a shipment amount is increased when the determined at least one first external factor occurs; And
When the occurrence of the at least one first external factor is predicted, the method of predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model, comprising the step of adjusting the amount of stocking of the first item and the second item. The method of claim 1,
Determining a stocking amount and a stocking time of the first item based on the predicted shipment quantity of the first item; And
In response to an irregular occurrence of at least one external factor, a method of predicting a delivery amount for each item of a distribution warehouse using an artificial intelligence model, comprising the step of modifying the stocking amount and the stocking time of the first item. The method of claim 1,
Generating the relationship graph,
Generating a plurality of nodes corresponding to a plurality of external factors occurring at the time of delivery of the first item;
Generating at least one line connecting each of the plurality of external factors to indicate a relationship between the plurality of external factors; And
And indicating the strength of a relationship between a plurality of external factors simultaneously occurring at the time of shipment of the first item by adjusting the thickness of the at least one line. The method of claim 1,
Generating first item arrangement information of the first distribution warehouse by using the location information and quantity information of the first item; And
And modifying the first item arrangement information in the distribution warehouse using the predicted delivery amount and the predicted delivery time of the first item. A memory for storing one or more instructions; And
A processor that executes the one or more instructions stored in the memory,
The processor executes the one or more instructions,
A device for predicting the amount of shipment for each item of a distribution warehouse using an artificial intelligence model that performs the method of claim 1. A computer program combined with a computer as hardware and stored in a recording medium readable by a computer to perform the method of claim 1.

Images