system

JP2026175198APending Publication Date: 2026-11-02SOFTBANK GROUP CORP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2025069988
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-04-21
Publication Date
2026-11-02

AI Technical Summary

Benefits of technology

【0005】 本発明に係る課題を解決するための手段は、過去の売上データ、販売データ、地域の客層データを入力として受け付けるデータ取得部と、前記データ取得部から取得した各店舗ごとの売上履歴、販売履歴、地域属性情報を統合し、生成系AIを用いて各店舗ごとおよび各商品ごとの需要を予測する需要予測部と、前記需要予測部によって算出された需要予測値に基づき、各店舗ごとおよび各商品ごとに最適な在庫発注量を算出し、店舗担当者または代理店担当者に対して発注提案情報を出力する発注提案部とを備えるシステムを構成することであり、前記需要予測部は、店舗ID、商品ID、月、過去売上数、過去販売数、地域年齢層、地域所得水準などの複数の属性情報を入力特徴量として生成系AIモデルに入力し、過去の実績データと地域特性を考慮した上で将来の需要数を高精度に予測し、前記発注提案部は、前記需要予測値と在庫状況、発注リードタイム、発注単位、欠品リスクおよび過剰在庫リスクを総合的に考慮して最適な発注数量を算出し、ユーザーインターフェースを介して担当者に具体的な発注提案を提示する構成とすることで、各店舗ごとに異なる需要特性や制約条件を反映した最適な在庫管理および発注業務の自動化と効率化を実現する手段である。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026175198000001_ABST
    Figure 2026175198000001_ABST
Patent Text Reader

Abstract

We provide the system. [Solution] The present invention aims to improve the efficiency and optimization of inventory management operations in stores and agencies. It comprises a demand forecasting unit that centrally collects and integrates diverse information such as past sales data, sales data, regional customer data, weather, events, competitors, promotions, and logistics on a server, and uses generative AI to predict demand for each store and each product with high accuracy; an order proposal unit that automatically generates optimal order proposals by comprehensively considering multiple constraints such as forecast results, inventory status, order lead time, logistics constraints, and cost factors; and a field information input unit that reflects feedback from on-site staff in real time. This system prevents stockouts and excess inventory, and simultaneously achieves optimal allocation of management resources and improved customer satisfaction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method executed by at least one processor, the method comprising the steps of: receiving a user utterance; adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot's character; encoding the prompt; and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0004] The problem that this invention aims to solve is that in multiple stores and agencies, various factors such as different customer bases, regional characteristics, best-selling products, seasonal fluctuations, promotional events, and competitive situations are intricately intertwined for each store. Traditional inventory ordering operations, which heavily relied on the experience and intuition of the person in charge, were unable to fully utilize the vast and diverse information such as past sales data, sales data, and regional customer data, resulting in frequent problems such as stockouts, excess inventory, increased inventory costs, lost opportunities, and inefficient ordering operations. In other words, conventional inventory management methods made it difficult to rationally and accurately calculate the optimal order quantity for each store. This led to serious management challenges, especially for businesses with multiple stores or handling a wide variety of products, such as increased workload for staff, reliance on individual decision-making, wasted resources, and decreased customer satisfaction. Furthermore, even when using advanced analysis and AI technology for demand forecasting, a system that automatically proposes optimal orders while simultaneously considering regional characteristics, seasonal fluctuations, and multiple constraints had not been sufficiently established. Therefore, there was a need for a practical solution that was tailored to the actual work on the ground. Therefore, the present invention aims to achieve optimal demand forecasting and ordering suggestions for each store and each product automatically and with high accuracy by utilizing generative AI and integrating and analyzing diverse information such as past sales data, sales data, and regional customer demographic data, thereby simultaneously achieving efficiency improvements in inventory management, prevention of stockouts and excess inventory, cost reduction, reduced burden on staff, optimal allocation of management resources, and improved customer satisfaction. [Means for solving the problem]

[0005] The means for solving the problems according to the present invention is to configure a system comprising: a data acquisition unit that accepts past sales data, sales data, and regional customer data as input; a demand forecasting unit that integrates the sales history, sales history, and regional attribute information for each store acquired from the data acquisition unit and forecasts the demand for each store and each product using a generative AI; and an order proposal unit that calculates the optimal inventory order quantity for each store and each product based on the demand forecast value calculated by the demand forecasting unit and outputs order proposal information to the store manager or agent manager, wherein the demand forecasting unit receives the store ID, product ID By inputting multiple attribute information such as D, month, past sales figures, past sales volume, regional age group, and regional income level as input features into a generative AI model, and accurately predicting future demand while considering past performance data and regional characteristics, the order proposal unit calculates the optimal order quantity by comprehensively considering the demand forecast value, inventory status, order lead time, order unit, stockout risk, and excess inventory risk, and presents specific order proposals to the person in charge via a user interface, this is a means of automating and streamlining optimal inventory management and ordering operations that reflect the different demand characteristics and constraints of each store. [Brief explanation of the drawing]

[0006] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0007] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0008] First, let's explain the terminology used in the following explanation.

[0009] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0010] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0011] In the following embodiments, the labeled storage is one or a plurality of non-volatile storage devices that store various programs, various parameters, and the like. Examples of the non-volatile storage device include a flash memory (SSD (Solid State Drive)), a magnetic disk (for example, a hard disk), or a magnetic tape and the like.

[0012] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F manages communication between a plurality of computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark) and the like.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, may be only B, or may be a combination of A and B. In addition, in the present specification, when three or more matters are expressed by being connected with "and / or", the same concept as that for "A and / or B" is applied.

[0014] [First Embodiment]

[0015] FIG. 1 illustrates an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] As shown in FIG. 1, the data processing system 10 includes a data processing apparatus 12 and a smart device 14. An example of the data processing apparatus 12 includes a server.

[0017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of the "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network), and the like.

[0018] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, the camera 42, and the communication I / F 44 are also connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A receives user input via contact of an indicator by detecting the contact of the indicator (for example, a pen, a finger, or the like). The microphone 38B receives voice-based user input by detecting the user's voice. A control unit 46A in the processor 46 transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, a specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0022] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0024] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0026] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0027] (Example 1)

[0028] The flow of a specific process in Example 1 will now be described. Each part of the system described below is implemented by a data processing device 12 and a smart device 14. The data processing device 12 will be referred to as a "server," and the smart device 14 will be referred to as a "terminal."

[0029] (Embodiments for carrying out the present invention)

[0030] Embodiments for carrying out the present invention will be described in more specific and detail. First, the system of the present invention consists of a server and a terminal, with a data acquisition unit, a demand forecasting unit and an order proposal unit on the server side, and a user interface unit and a field information input unit on the terminal side.

[0031] The data acquisition unit operates on a server and receives and stores sales data, sales data, inventory data, regional customer data, weather data, traffic information, event information, competitor store information, promotional information, and logistics information in various formats from each store and agency. For example, sales data includes daily, weekly, and monthly sales quantities, sales amounts, sales by product category, sales by time of day, sales by day of the week, and sales during specific events. Sales data includes sales quantities, unit prices, discount history, and campaign application history for each product. Inventory data includes inventory quantities for each store, receiving dates, shipping dates, expiration dates, disposal history, return history, and inventory turnover rate. Regional customer data includes demographic data around the store, age distribution, household composition, income levels, housing types, commuting patterns, history of local events, and local purchasing characteristics. Weather data includes temperature, precipitation, humidity, weather changes, and abnormal weather information such as typhoons and heavy snow. Traffic information includes the number of users at the nearest train station and bus stop, road congestion, delivery time from logistics centers to stores, and delay information due to traffic accidents and construction. Event information includes local festivals, sports competitions, concerts, exhibitions, openings and renovations of commercial facilities, school entrance and graduation ceremonies, and seasonal events. Competitor store information includes new store openings and closures in the vicinity, competitor store promotional strategies, price trends, product assortment information, and changes in business hours. Promotional information includes the store's own campaigns, point rewards, coupon distribution, advertising, social media announcements, and data measuring the effectiveness of past promotional strategies. Logistics information includes order lead time, delivery frequency, delivery costs, logistics capacity, delivery delay history, and the operational status of logistics companies. This data is collected on the server from various sources, including POS systems, inventory management systems, customer management systems, external statistical databases, open data APIs, IoT sensors, and terminal input from on-site staff.

[0032] The demand forecasting unit operates on a server and performs demand forecasting using generative AI. The generative AI is composed of a combination of, for example, large-scale language models, time series forecasting models, deep learning models, reinforcement learning models, and transformer models. The generative AI comprehensively preprocesses multidimensional and multitemporal data obtained from the data acquisition unit, performing tasks such as missing value imputation, outlier removal, feature engineering, lag feature generation for time series data, categorical data encoding, and text data vectorization. The generative AI makes extremely specific predictions, such as "how much of product X will be sold at store A next month," "how much demand for product Y will there be at store B during the summer," "what initial demand for newly introduced product Z can be expected at store C," "how much will demand for a specific product increase at store D when an event is held," and "how much will demand for beverages fluctuate at store E when the weather deteriorates." Generative AI learns by relating diverse factors such as past sales, sales, inventory, customer demographics, weather, events, competitors, promotions, and logistics in a time-series and spatial manner. For example, it automatically extracts and learns things like "sales trends for the same period over the past three years," "promotional effects over the past month," "regional population growth / decrease trends," "impact of new store openings by competitors," and "demand fluctuation patterns during sudden weather changes." Generative AI enables demand forecasting that reflects regional and seasonal characteristics, such as "Store A has a large elderly population, so there is high demand for health foods," "Store B has a high income level, so demand for high-priced products is stable," "Store C has a large student population, so demand for low-priced products fluctuates greatly seasonally," "Store D sees a surge in demand for beverages and snacks when events are held," and "Store E sees increased demand for preserved foods and daily necessities when the weather deteriorates." Generative AI can also simultaneously output factor analysis and highly explanatory forecasting evidence, such as "why demand for this product increases at this time," "which factors contribute to demand fluctuations," and "which factors are most important at which store."

[0033] The order proposal unit operates on a server and calculates the optimal order quantity by comprehensively considering multiple constraints, such as inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, past ordering trends of the person in charge, order cycle for each store, and the operating status of logistics companies, based on the demand forecast values ​​for each store and each product output from the demand forecasting unit. For example, if a sudden increase in demand for product X is predicted at store A, the optimal order quantity is calculated by taking into account the current inventory level, the number of days from order to delivery, the minimum order unit, available warehouse space, logistics congestion, past stockout history, order cost, inventory cost, and waste risk. To solve the optimization problem of these complex constraints, the generative AI generates order proposals by combining methods such as reinforcement learning, evolutionary algorithms, mathematical optimization techniques, and simulation optimization. Furthermore, the ordering proposal department can also perform scenario analysis and risk prediction using generative AI, such as "Demand is likely to increase next week due to a forecast of worsening weather this week," "Demand for a specific product will temporarily increase due to a large-scale event being held nearby," "The logistics company's operations will be restricted, so the timing of orders needs to be brought forward," or "Inventory space is running low, so the order quantity needs to be reduced," and present multiple ordering scenarios to the person in charge. For example, it can simultaneously present ordering proposals for multiple cases such as a normal scenario, a bad weather scenario, an event scenario, and a logistics delay scenario, enabling the person in charge to make the optimal ordering decision according to the situation.

[0034] On the terminal side, store staff and agency staff receive order proposal information distributed from the server and perform operations such as confirmation, approval, modification, and order execution through the user interface. For example, specific proposals such as "The recommended order quantity for product X is 50 units," "Based on past order trends, demand is expected to increase this week," "Due to inventory space constraints, the maximum order quantity is 80 units," and "Considering the risk of increased demand due to bad weather, the recommended order quantity is 60 units" are displayed on the tablet screen, and staff can use this as a reference to decide on the order quantity. Furthermore, staff can input new information obtained on-site from the terminal side, such as "A nearby store has closed," "A new competing store has opened," "A new event is scheduled in the area," "We received notification of a delivery delay from the logistics company," or "Warehouse space has been temporarily expanded," and reflect this in real time in the server-side generative AI. For example, if a staff member inputs from the terminal that "A large-scale local event will be held this weekend," the server-side generative AI can immediately reflect this information in the demand forecasting model, automatically generate a new order proposal that takes the event demand into account, and redistribute it to the terminal.

[0035] For example, if store A has a large elderly customer base and high demand for health foods, the generative AI combines regional age group data with past sales data to predict increased demand for health foods, and the ordering department calculates the order quantity that reflects this. For example, if store B has a high income level and high-priced products sell well, the generative AI combines income level data with product unit price data to predict demand for high-priced products, and the ordering department makes ordering suggestions based on this. For example, if store C has products whose demand fluctuates greatly from season to season, the generative AI combines past seasonal fluctuation data with weather data to predict seasonal demand fluctuations, and the ordering department proposes the optimal ordering timing and quantity. For example, if store D is implementing a sales promotion campaign or point reward program, the generative AI combines promotional information with past sales data from promotions to predict increased demand during the promotion period, and the ordering department calculates the order quantity that reflects this. For example, if store E is experiencing logistics congestion or fluctuations in ordering lead times, the generative AI considers logistics data and past delivery delay data to propose ordering timing and quantity that minimizes the risk of delivery delays. For example, if an employee's past ordering trends or on-site judgments have a specific bias, the generative AI will analyze each employee's ordering history, correct for the bias, and then propose the most suitable order.

[0036] Furthermore, for example, when store F introduces a new product, the generative AI comprehensively analyzes past data of similar products, regional characteristics, sales promotion plans, and competitive situations to predict initial demand for the new product, and the ordering proposal department proposes the initial order quantity based on this. For example, when store G handles fresh food with a high risk of waste, the generative AI takes into account expiration date data, past waste history, temperature and humidity data, inventory turnover rate, etc., to calculate the order quantity that minimizes waste risk. For example, when logistics costs are soaring at store H, the generative AI comprehensively optimizes logistics cost data, order units, delivery frequency, inventory space, etc., to propose cost-effective ordering. For example, when inventory can be transferred between multiple stores at store I, the generative AI comprehensively optimizes the inventory status of each store, demand forecasts, logistics costs, transferable quantities, etc., to propose ordering and replenishment including inter-store inventory transfers.

[0037] In this embodiment, the server-side implements a data acquisition unit, a demand forecasting unit, and an order proposal unit. By using a generative AI to comprehensively analyze, learn, predict, and optimize diverse and vast amounts of data, the system automatically and accurately generates optimal order proposals for each store and each product. On the terminal side, the person in charge can confirm, modify, and execute the proposal, and receive real-time feedback from the field. This configuration simultaneously achieves increased efficiency in inventory management, prevention of stockouts and excess inventory, cost reduction, optimal allocation of management resources, and improved customer satisfaction.

[0038] (System Configuration)

[0039] The system according to this embodiment comprises a data acquisition unit, a demand forecasting unit, an order proposal unit, a user interface unit, a field information input unit, and a generation-based AI collaboration unit. The data acquisition unit has the function of receiving sales data, sales data, inventory data, regional customer demographics data, weather data, traffic information, event information, competitor store information, promotional information, logistics information, etc., transmitted from each store and agency in various formats and storing them in a database on the server. For example, sales data includes daily, weekly, and monthly sales quantities, sales amounts, sales by product category, sales by time of day, sales by day of the week, and sales during specific events. Sales data includes sales quantities, unit prices, discount history, and campaign application history for each product. Furthermore, inventory data manages inventory quantities for each store, receiving dates, shipping dates, expiration dates, disposal history, return history, and inventory turnover rate. Regional customer demographics data includes demographic data around the store, age distribution, household composition, income levels, housing types, commuting patterns, history of local events, and regional purchasing characteristics. Weather data includes temperature, precipitation, humidity, weather changes, and abnormal weather information such as typhoons and heavy snow. Traffic information includes the number of users at the nearest train station and bus stop, road congestion, delivery time from logistics centers to stores, and delays due to traffic accidents and construction. Event information includes local festivals, sports competitions, concerts, exhibitions, openings and renovations of commercial facilities, school entrance and graduation ceremonies, and seasonal events. Competitor store information includes new store openings and closures in the vicinity, promotional strategies of competitors, price trends, product assortment information, and changes in business hours. Promotional information includes the store's own campaigns, point rewards, coupon distribution, advertising, SNS announcements, and data on the effectiveness of past promotional strategies. Logistics information manages order lead times, delivery frequency, delivery costs, logistics capacity, delivery delay history, and the operational status of logistics companies. This data is aggregated on the server from various sources, including POS systems, inventory management systems, customer management systems, external statistical databases, open data APIs, IoT sensors, and terminal input from on-site staff.

[0040] The demand forecasting unit has the function of comprehensively preprocessing multidimensional and multitemporal data obtained from the data acquisition unit, inputting the data into the generative AI via the generative AI collaboration unit, and performing demand forecasting. For example, the demand forecasting unit performs preprocessing such as missing value imputation, outlier removal, feature engineering, lag feature generation for time series data, encoding of categorical data, and vectorization of text data, and sends specific forecast instructions as prompts to the generative AI, such as "how much of product X will be sold at store A next month," "how much demand for product Y will there be at store B during the summer," "how much initial demand can be expected for the newly introduced product Z at store C," "how much will the demand for a specific product increase when an event is held at store D," and "how much will the demand for beverages fluctuate at store E when the weather deteriorates." The generative AI learns by relating various factors such as past sales, sales, inventory, customer demographics, weather, events, competitors, promotions, and logistics both chronologically and spatially. For example, it automatically extracts and learns "sales trends for the same period over the past three years," "promotional effects over the past month," "regional population growth / decrease trends," "impact of new store openings by competitors," and "demand fluctuation patterns during sudden weather changes," and returns the prediction results to the demand forecasting department. The demand forecasting department aggregates the prediction values ​​obtained from the generative AI for each store and each product and sends them to the order proposal department.

[0041] The Order Proposal Department has the function of calculating the optimal order quantity by comprehensively considering multiple constraints, such as inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, past ordering trends of the person in charge, order cycle for each store, and the operating status of logistics companies, based on the demand forecast values ​​for each store and each product output from the Demand Forecasting Department. For example, if the Order Proposal Department predicts a sudden increase in demand for product X at store A, it will calculate the optimal order quantity by taking into account the current inventory level, the number of days from order to delivery, the minimum order unit, available warehouse space, logistics congestion, past stockout history, order cost, inventory cost, and waste risk. Furthermore, the ordering proposal department sends scenario analyses and risk predictions to the generation AI as prompts via the generation AI collaboration department, such as "Demand is likely to increase next week due to forecasts of worsening weather this week," "Demand for a specific product will temporarily increase due to a large-scale event being held nearby," "The logistics company's operations will be restricted, so the timing of orders needs to be brought forward," and "Inventory space is running low, so the order quantity needs to be reduced." The AI ​​then generates multiple ordering scenarios simultaneously and presents them to the person in charge. For example, it simultaneously presents ordering proposals for multiple cases, such as a normal scenario, a bad weather scenario, an event scenario, and a logistics delay scenario, enabling the person in charge to make the optimal ordering decision according to the situation.

[0042] The user interface unit receives order suggestion information distributed from the server to store and agency staff and displays specific suggestions on the screen, such as "The recommended order quantity for product X is 50 units," "Based on past ordering trends, demand is expected to increase this week," "Due to inventory space constraints, the maximum order quantity is 80 units," and "Considering the risk of increased demand due to bad weather, the recommended order quantity is 60 units," allowing staff to decide on the order quantity based on this information. Furthermore, the user interface unit also includes input screens for staff to review, approve, modify, and execute orders, as well as a dashboard function that visualizes past ordering history and the results of factor analysis that form the basis of demand forecasts. For example, if a staff member enters that "a large-scale local event will be held this weekend," the user interface unit can send this information to the field information input unit, which can then be reflected in the server-side generation AI in real time.

[0043] The field information input unit has the function of inputting new information obtained by the person in charge on-site, such as "a nearby store has closed," "a new competing store has opened," "a new event is scheduled in the area," "a delivery delay has been reported by the logistics company," or "warehouse space has been temporarily expanded," and reflecting it in real time to the server-side generation AI. The field information input unit immediately transmits the input information to the data acquisition unit and the demand forecasting unit, so that the generation AI can recalculate demand forecasts and order proposals that take the latest field information into account. For example, if the person in charge inputs that "a large-scale event will be held in the area this weekend," the generation AI can immediately reflect this information in the demand forecasting model, automatically generate new order proposals that take the event demand into account, and redistribute them to the terminal.

[0044] The Generative AI Integration Unit operates on a server and has the function of sending instructions and data from the Demand Forecasting Unit and the Order Proposal Unit to the Generative AI in an appropriate prompt format, and receiving the output results from the Generative AI and returning them to each unit. The Generative AI Integration Unit generates specific prompt sentences such as, "Forecast the demand for product X at store A next month. Input data includes sales data, sales data, inventory data, regional age distribution, income level, weather data, event information, competitor store information, promotional information, and logistics information for the past three years," "Forecast the summer demand for product Y at store B and propose the optimal order quantity considering inventory space and order lead time," "Forecast the initial demand for newly introduced product Z at store C based on past data of similar products and regional characteristics," "Forecast how much the demand for a specific product will increase at store D when an event is held, based on past event sales data and regional event information," "Forecast how much the demand for beverages will fluctuate at store E when the weather deteriorates, based on past weather data and sales data," and "Optimize the timing and quantity of orders when the logistics provider's operations are restricted," and sends them to the Generative AI. The Generative AI Integration Unit returns demand forecasts, order proposals, factor analysis results, and scenario analysis results returned from the Generative AI to the Demand Forecasting Unit and the Order Proposal Unit, thereby achieving advanced automation and optimization of the entire system.

[0045] As described above, the system according to this embodiment includes a data acquisition unit, a demand forecasting unit, an order proposal unit, a user interface unit, a field information input unit, and a generation AI collaboration unit. Each unit works in cooperation to comprehensively analyze, learn, predict, and optimize diverse and vast amounts of data, automatically and accurately realizing optimal order proposals for each store and each product. The system is configured to allow field personnel to confirm, modify, and execute the proposals, and to provide real-time feedback of field information. This configuration simultaneously achieves increased efficiency in inventory management, prevention of stockouts and excess inventory, cost reduction, optimal allocation of management resources, and improved customer satisfaction.

[0046] (Implementation Steps)

[0047] Step 1: Data Collection and Integration A wide range of data, including sales data, sales data, inventory data, regional customer data, weather data, traffic information, event information, competitor store information, promotional information, and logistics information, is centrally collected and integrated by the data acquisition unit on the server. For example, sales data includes daily, weekly, and monthly sales quantities and sales amounts, sales by product category, sales by time of day, and sales during specific events. Inventory data includes inventory quantity, receiving date, shipping date, expiration date, disposal history, return history, and inventory turnover rate. Regional customer data includes demographic data, age distribution, household composition, income level, housing type, commuting / schooling patterns, history of local events, and purchasing characteristics. This data is automatically or manually aggregated from various sources, such as POS systems, inventory management systems, external statistical databases, open data APIs, IoT sensors, and terminal input from field staff.

[0048] Step 2: Data preprocessing and feature generation The multidimensional, multitemporal data collected by the data acquisition unit undergoes preprocessing on the server, including missing value imputation, outlier removal, feature engineering, lag feature generation for time series data, categorical data encoding, and text data vectorization. For example, if historical sales data has missing values, imputation is performed using statistical methods or machine learning, and outliers are removed using business rules or statistical thresholds. Feature engineering generates new features such as the average sales for the same period over the past three years, the effectiveness of promotions in the most recent month, regional population growth / decrease trends, the impact of new store openings by competitors, and demand fluctuation patterns during sudden weather changes.

[0049] Step 3: Demand forecasting using generative AI Pre-processed data is input into the generating AI, and demand forecasts are performed for each store and each product. The generating AI integration unit generates prompt messages, such as the following, and sends them to the generating AI. Specific examples of prompt messages: "Forecast the demand for product X at store A next month. Input data includes sales data, sales data, inventory data, regional age distribution, income levels, weather data, event information, competitor store information, promotional information, and logistics information for the past three years." "Forecast the summer demand for product Y at store B and propose the optimal order quantity considering inventory space and order lead time." "Forecast the initial demand for newly introduced product Z at store C based on past data of similar products and regional characteristics." "Forecast how much the demand for a specific product will increase at store D during an event, based on past event sales data and regional event information." Based on these prompts, the generating AI comprehensively analyzes time-series and spatial factors and outputs highly accurate demand forecasts for each store and each product.

[0050] Step 4: Order Optimization and Order Proposal Generation Based on the demand forecast values ​​for each store and product output from the demand forecasting department, the ordering proposal department calculates the optimal order quantity by comprehensively considering multiple constraints such as inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, past ordering trends of the person in charge, ordering cycle for each store, and the operating status of logistics companies. The ordering proposal department sends scenario analyses and risk predictions to the generation AI as prompts via the generation AI collaboration department, such as "Demand is likely to increase next week because the weather is forecast to worsen this week," "Demand for a specific product will temporarily increase because a large-scale event is being held nearby," "The logistics company's operations are restricted, so the ordering timing needs to be brought forward," and "Inventory space is tight, so the order quantity needs to be reduced." The AI ​​then generates multiple ordering scenarios simultaneously and presents them to the person in charge.

[0051] Step 5: Proposal presentation and on-site information input via user interface The user interface displays order suggestion information delivered from the server to store and agency representatives on the screen, presenting specific suggestions such as "The recommended order quantity for product X is 50 units," "Based on past ordering trends, demand is expected to increase this week," "Due to inventory space constraints, the maximum order quantity is 80 units," and "Considering the risk of increased demand due to bad weather, the recommended order quantity is 60 units." The representatives use this information to determine the order quantity and input any new information obtained on-site, such as "A nearby store has closed," "A new competing store has opened," "A new event is planned in the area," "We received notification of a delivery delay from the logistics company," or "Warehouse space has been temporarily expanded," through the on-site information input unit. This on-site information is immediately reflected in the server-side AI generation system, and demand forecasts and order suggestions are recalculated in real time.

[0052] Step 6: Feedback and Continuous Learning After an order is placed, the data acquisition unit collects actual sales data, inventory consumption data, waste data, stockout status, order history, and feedback information from the field, and stores them as training data for the generating AI. The generating AI uses this historical data to continuously improve the accuracy of the model, correct biases, and refine factor analysis, thereby improving the accuracy of future demand forecasts and order proposals. For example, it automatically analyzes factors such as "what caused the previous order proposal to result in excess inventory," "what caused stockouts," and "the impact of promotional measures on demand," in order to optimize the entire system.

[0053] (Specific use cases)

[0054] For example, let's consider the application of the present invention's system to inventory management and ordering operations for a product group mainly consisting of beverages and snacks at a store located in front of a train station in an urban area. The data acquisition unit acquires daily sales data, hourly sales data, sales data by product category, inventory quantity, receiving date, shipping date, expiration date, disposal history, return history, inventory turnover rate, as well as demographic data of the area around the store, including age distribution, household composition, income level, housing type, commuting / schooling patterns, history of local events, purchasing characteristics, temperature, precipitation, humidity, weather changes, abnormal weather information such as typhoons and heavy snow, number of users at the nearest train station or bus stop, road congestion status, delivery time from logistics base to store, and information on traffic accidents and construction. The system centrally collects and integrates information in various formats, including delay information, local festivals, sports competitions, concerts, exhibitions, openings and renovations of commercial facilities, school entrance and graduation ceremonies, seasonal events, new store openings and closures, competitor promotional strategies, price trends, product assortment information, changes in business hours, the store's own campaigns, point rewards, coupon distribution, advertising placements, social media announcements, data on the effectiveness of past promotional strategies, order lead times, delivery frequency, delivery costs, logistics capacity, delivery delay history, and logistics provider operational status.

[0055] The demand forecasting unit performs preprocessing on this multidimensional, multi-timescale data, including missing value imputation, outlier removal, feature engineering, lag feature generation for time-series data, categorical data encoding, and text data vectorization, and then sends prompt messages to the generating AI via the generating AI collaboration unit. For example, it can generate prompt messages like the following: The prompt will generate and send to the AI ​​prompts such as: "Store ID: 101, Product Category: Beverages, Period: Next month, Input data: Daily sales data for the past 3 years, sales data by time of day, inventory data, regional demographic data, age distribution, income level, weather data, regional event information, competitor store information, promotional information, logistics information. Based on the above data, calculate the forecast demand for the entire beverage category and for each major product next month, and output the importance of the factors influencing demand fluctuations." For demand forecasts during specific events, the prompt will generate and send to the generating AI prompts such as: "Store ID: 101, Product ID: A001, Event: Regional Summer Festival, Period: Second week of August, Input data: Past sales data for the same event, weather data, regional demographics, competitor store promotional information, logistics information. Based on the above data, calculate the forecast demand and stockout risk for the relevant product during the event period."

[0056] Based on these prompts, the generating AI comprehensively analyzes time-series and spatial factors, automatically extracting and learning elements such as "sales trends for the same period over the past three years," "promotional effects over the past month," "regional population growth / decrease trends," "impact of new store openings by competitors," and "demand fluctuation patterns during sudden weather changes." It then outputs highly accurate demand forecasts and the importance of demand fluctuation factors for each store and product.

[0057] The ordering proposal department calculates the optimal order quantity by comprehensively considering multiple constraints, such as inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, ordering cost, inventory cost, waste risk, past ordering trends of the person in charge, ordering cycle for each store, and the operating status of logistics providers, based on the demand forecast values ​​for each store and each product output from the demand forecasting department. For example, under the condition that "For product A, a major beverage item, next month's demand forecast is 500 units, current inventory is 100 units, order lead time is 3 days, minimum order unit is 50 units, warehouse space limit is 600 units, logistics provider delivery frequency is twice a week, past stockout history is 2 times in the last 6 months, calculate the order quantity that minimizes ordering and inventory costs," the order proposal unit sends the prompt message "For product A, next month's demand forecast is 500 units, current inventory is 100 units, order lead time is 3 days, minimum order unit is 50 units, warehouse space limit is 600 units, logistics provider delivery frequency is twice a week, past stockout history is 2 times in the last 6 months, propose the order quantity that minimizes ordering and inventory costs" to the generation AI integration unit, which then outputs the optimal order quantity and the underlying factor analysis.

[0058] The user interface displays order suggestion information delivered from the server to store and agency representatives on the screen, presenting specific suggestions such as, "The recommended order quantity for product A is 400 units. Considering past ordering trends and demand fluctuation factors, demand is expected to increase this month due to a local event. Due to warehouse space constraints, the maximum order quantity is 500 units. Considering the delivery frequency of the logistics company, it is also possible to place the order in two installments." The representative uses this information to determine the order quantity and inputs any new information obtained on-site, such as "A nearby store has closed," "A new competing store has opened," "A new event is scheduled in the area," "We received notification of a delivery delay from the logistics company," or "Warehouse space has been temporarily expanded," through the on-site information input section. This on-site information is immediately reflected in the server-side AI generation system, and demand forecasts and order suggestions are recalculated in real time.

[0059] Thus, the system of the present invention, in the inventory management and ordering operations of beverages and snacks at stores in front of train stations in urban areas, integrates and analyzes various factors such as past sales data, inventory data, regional characteristics, weather, events, competitors, promotions, and logistics. It automatically achieves highly accurate demand forecasting and optimal ordering suggestions using generated AI, and by reflecting on-site information feedback from staff in real time, it is possible to simultaneously achieve efficiency in inventory management operations, prevention of stockouts and excess inventory, cost reduction, optimal allocation of management resources, and improved customer satisfaction.

[0060] (Application Example 1) The flow of a specific process in Application Example 1 will be explained below. Each part of the system described below is implemented by a data processing device 12 and a smart device 14. The data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0061] (Embodiments for carrying out the present invention)

[0062] Embodiments for carrying out the present invention will be described in more detail and specifically, including the system configuration and operation examples in a logistics center. The system of this embodiment consists of a server and multiple terminals. The server is equipped with an inbound / outbound data acquisition unit, a demand forecasting unit, an inventory optimization proposal unit, a generation AI integration unit, a database management unit, and an external information integration unit. The terminals are equipped with a field input unit, a user interface unit, and a notification / alert unit.

[0063] The Inbound / Outbound Data Acquisition Unit automatically collects and stores basic inventory data for all products handled at the logistics center, including the number of incoming and outbound items, inventory quantity, incoming date, outbound date, expiration date, disposal history, return history, inventory turnover rate, lot number, storage area, and temperature control information. For example, it can record product information in real time during inbound and outbound shipments using barcode scanners and RFID readers and transmit it to the server. Furthermore, during inbound shipments, it simultaneously records supplier information, inbound inspection results, defect rate, and storage area allocation status. During outbound shipments, it also records delivery destination information, outbound inspection results, picking personnel, outbound time, and delivery route information for the outbound destination. For products requiring expiration date management, expiration date approaching alerts and disposal schedule lists are automatically generated. For return history, details such as return reason, returned quantity, return destination, and whether re-inbound is possible are also recorded. Storage area information includes storage conditions such as refrigeration, freezing, ambient temperature, high humidity, and low humidity, as well as shelf numbers, pallet numbers, storage capacity, and temperature and humidity history of the storage area.

[0064] The External Information Integration Department automatically acquires and integrates diverse external data, including demand indices for each delivery destination region, weather information, traffic congestion levels, event impact levels, number of workers, worker shifts, delivery performance data, delivery destination characteristics, vehicle operating status, worker skills and health status, logistics provider operating status, external warehouse inventory status, and operating status of competing logistics centers. For example, demand indices are automatically acquired from external statistical databases and past delivery performance data, tailored to regional characteristics such as urban areas, suburbs, tourist destinations, and industrial zones. Weather information, such as temperature, precipitation, snowfall, typhoons, heat waves, cold waves, and sudden weather changes, is acquired in real time from weather APIs and IoT sensors. Traffic congestion levels are acquired from traffic information APIs and GPS data, including congestion on major roads, construction and accident information on logistics routes, real-time location information of delivery vehicles, and road closure information. Event impact levels are automatically collected from local event calendars and social media information, including local festivals, sports competitions, concerts, exhibitions, openings and renovations of commercial facilities, school holidays, sales information, and social media buzz. Worker numbers and shift information are obtained from the human resources management system and wearable devices, including daily attendance schedules, worker skill levels, staffing levels during peak seasons, worker health status and rest periods, worker qualifications, and worker past work performance. Delivery performance data is obtained from the delivery management system, including delivery history for each destination, delivery delay history, delivery vehicle operating status, delivery time for each delivery route, delivery costs, delivery vehicle fuel consumption, and delivery vehicle maintenance history. Inventory status at external warehouses and operating status of competing logistics centers are also obtained through external APIs and EDI systems.

[0065] The Database Management Department manages this diverse data from multiple perspectives, including by time series, product, region, worker, vehicle, and storage area, ensuring data integrity, consistency, and security. For example, the database consists of multiple tables, such as an inbound / outbound history table, inventory table, expiration date table, return table, delivery performance table, worker shift table, vehicle operation table, and external information table, and strictly manages the relationships between each table.

[0066] The demand forecasting unit preprocesses multidimensional, multi-temporal data acquired from the inventory data acquisition unit, the external information linkage unit, and the database management unit, and inputs it into the generative AI via the generative AI linkage unit. Preprocessing includes missing value imputation, outlier removal, feature engineering, lag feature generation for time series data, categorical data encoding, text data vectorization, anomaly detection, and data normalization. For example, it extracts complex cases as features from past inventory inflows, outflows, and inventory trends, such as "beverage inventory plummeted when events coincided just before a public holiday and the weather deteriorated," "outflow delays occurred on days when there was a shortage of workers," "waste losses increased during weeks when the expiration dates of refrigerated products were approaching," and "delivery delays occurred frequently on days when delivery vehicles broke down." The generative AI receives prompts such as, "Product ID: A001, Period: Next week, Input data: Inbound / outbound data for the past 3 years, inventory data, regional demand index, weather data, traffic congestion level, event impact level, number of workers, delivery performance data, expiration date data, storage area information. Based on the above data, calculate the optimal inventory level and the importance of demand fluctuation factors for next week." Based on these prompts, the generative AI comprehensively analyzes time-series and spatial factors and outputs the optimal inventory level and the importance of demand fluctuation factors for each product with high accuracy. For example, it might say, "The event impact level is high and the demand index is also high, so you should secure more inventory than usual," "The weather index is low, so there is a risk of delivery delays, and you need ample inventory," "There are enough workers, so there is ample capacity for inbound / outbound processing," or "There are many products nearing their expiration date, so we recommend ordering with the risk of waste in mind," and then comprehensively judge multiple factors to suggest the optimal inventory level.

[0067] The Inventory Optimization Proposal Department comprehensively considers multiple constraints, such as the demand forecast value for each product output from the Demand Forecasting Department, the current inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, worker utilization status, delivery vehicle utilization status, delivery plan, workload, storage area availability, inventory status of external warehouses, and the possibility of inventory transfer between multiple locations. Using generative AI, it automatically generates the optimal inventory quantity, inbound and outbound timing, delivery plan, workload distribution plan, and inventory transfer plan for each product. For example, under the conditions "For product A, the demand forecast value for next week is 500 units, the current inventory is 100 units, the order lead time is 3 days, the minimum order unit is 50 units, the warehouse space limit is 600 units, the logistics provider delivers twice a week, there has been a stockout history of 2 times in the last 6 months, calculate the order quantity that minimizes order cost and inventory cost," it calculates the optimal order quantity and inbound timing. Furthermore, in cases of high event impact or low weather index, we propose securing ample inventory and accelerating delivery plans, taking into account the risks of delivery delays and sudden increases in demand. If there is a shortage of workers, we automatically generate an inbound / outbound schedule that equalizes the workload, taking into account the constraints of inbound / outbound processing capacity. We also propose inventory sharing between multiple locations and optimization of delivery routes, depending on the operating status of delivery vehicles and the congestion of delivery routes. For example, by sharing inventory between multiple logistics centers, we can achieve an overall optimal inventory allocation and improve delivery efficiency. For products requiring temperature control, such as refrigerated goods and pharmaceuticals, we propose inventory management that minimizes quality risks, taking into account the temperature and humidity history of the storage area and the operating status of refrigerated vehicles.

[0068] The on-site input unit allows on-site personnel at the logistics center to input real-time information that occurs on-site, such as "a sudden large order from a delivery destination," "an unexpected worker absence," "an accident or equipment failure in the warehouse," "a decision to hold a new local event," "notification of a delivery delay from a logistics company," "an urgent request to move inventory from an external warehouse," or "temperature deviation due to a refrigeration equipment failure," via a terminal. This information can then be immediately reflected in the server-side generation AI. As a result, demand forecasts and inventory optimization suggestions are recalculated in real time based on the latest situation on-site.

[0069] The user interface displays operational suggestions on the screen for logistics center operators and on-site staff, including optimal inventory levels, inbound and outbound timing, delivery plans, workload distribution proposals, inventory transfer proposals, expiration date alerts, disposal schedules, worker shift adjustment proposals, and delivery vehicle operation plans, all distributed from the server. For example, it presents specific and practical suggestions from multiple perspectives, such as "The recommended inventory level for product A is 400 units, we recommend two inbound shipments this week, and plan additional deliveries before the event," "On Wednesdays when there is a shortage of workers, we will reduce inbound and outbound operations and concentrate them on Thursday," "On Fridays when traffic congestion is expected, we will change the delivery route and recommend early morning deliveries," "Since there are many refrigerated products nearing their expiration dates, we will conduct a sales promotion campaign this week to accelerate inventory consumption," and "We will transfer inventory from external warehouses on Monday to prepare for a surge in demand on Tuesday." Staff members use this information to make operational decisions and can continuously optimize the entire system by inputting additional on-site information as needed.

[0070] The notification and alert unit detects operational risks and anomalies in real time, such as approaching expiration dates, insufficient stock, excess stock, delivery delays, labor shortages, equipment failures, urgent requests from external warehouses, traffic disruptions, and sudden weather changes, and immediately notifies the staff's terminals and the administrator's email / messaging system. For example, it can detect situations such as "Stock of product B is expected to run out today," "Temperature deviation detected in refrigeration equipment," "Delivery delays due to delivery vehicle breakdown," "Shift readjustment required due to sudden worker absence," and "Emergency stock transfer request received from external warehouse," thereby enhancing on-site responsiveness.

[0071] In this embodiment, the server side implements an inventory data acquisition unit, a demand forecasting unit, an inventory optimization proposal unit, a generation AI integration unit, a database management unit, and an external information integration unit, while the terminal side implements a field input unit, a user interface unit, and a notification / alert unit. This enables the realization of an advanced operational support system that optimizes inventory management, inventory planning, delivery planning, workload distribution, quality control, risk management, and external integration in a logistics center from multiple perspectives and in real time, simultaneously achieving prevention of stockouts and excess inventory, cost reduction, improved work efficiency, reduced quality risk, improved customer satisfaction, and improved on-site responsiveness.

[0072] (System Configuration)

[0073] The system according to this embodiment includes an inventory data acquisition unit, an external information linkage unit, a database management unit, a demand forecasting unit, an inventory optimization proposal unit, a generation AI linkage unit, a field input unit, a user interface unit, and a notification / alert unit. The Inbound / Outbound Data Acquisition Unit automatically collects and stores information on all products in the logistics center, including the number of incoming and outbound items, inventory quantity, incoming date, outbound date, expiration date, disposal history, return history, inventory turnover rate, lot number, storage area, and temperature control information. For example, it can record product information in real time during inbound and outbound shipments using barcode scanners and RFID readers and transmit it to the server. During inbound shipments, it also records supplier information, inbound inspection results, defect rate, and storage area allocation status. During outbound shipments, it records delivery destination information, outbound inspection results, picking personnel, outbound time, and delivery route information for the outbound destination. For products requiring expiration date management, it automatically generates expiration date alerts and disposal schedule lists. For return history, it records details such as return reason, returned quantity, return destination, and whether re-inbound is possible. It also manages storage area information such as storage conditions (refrigerated, frozen, ambient temperature, high humidity, low humidity), shelf number, pallet number, storage capacity, and temperature and humidity history of the storage area.The External Information Integration Department automatically acquires and integrates diverse external data, such as demand indices for each delivery destination region, weather information, traffic congestion levels, event impact levels, number of workers, worker shifts, delivery performance data, delivery destination characteristics, vehicle operating status, worker skills and health status, logistics provider operating status, external warehouse inventory status, and operating status of competing logistics centers. For example, demand indices are automatically acquired from external statistical databases and past delivery performance data, tailored to regional characteristics such as urban areas, suburbs, tourist destinations, and industrial zones. Weather information, including temperature, precipitation, snowfall, typhoons, heat waves, cold waves, and sudden weather changes, is acquired in real time from weather APIs and IoT sensors. Traffic congestion levels are acquired from traffic information APIs and GPS data, including congestion on major roads, construction and accident information on logistics routes, real-time location information of delivery vehicles, and road closure information. The system acquires event impact data, such as local festivals, sports competitions, concerts, exhibitions, openings and renovations of commercial facilities, school holidays, sales information, and social media buzz, automatically collected from local event calendars and social media information. Worker numbers and shift information, such as daily attendance schedules, worker skill levels, staffing levels during peak seasons, worker health and rest status, worker qualifications, and worker past work performance, are acquired from human resources management systems and wearable devices. Delivery performance data, such as delivery history for each destination, delivery delay history, delivery vehicle operating status, time required for each delivery route, delivery costs, delivery vehicle fuel consumption, and delivery vehicle maintenance history, are acquired from the delivery management system. Inventory status at external warehouses and operating status of competing logistics centers are also acquired through external APIs and EDI systems. The Database Management Department manages this diverse data from multiple perspectives, including by time series, product, region, worker, vehicle, and storage area, ensuring data integrity, consistency, and security. It consists of multiple tables, such as a receiving / shipping history table, inventory table, expiration date table, return table, delivery performance table, worker shift table, vehicle operation table, and external information table, and strictly manages the relationships between each table.The demand forecasting unit preprocesses multidimensional, multi-temporal data acquired from the inventory data acquisition unit, the external information linkage unit, and the database management unit, and inputs it into the generative AI via the generative AI linkage unit. Preprocessing includes missing value imputation, outlier removal, feature engineering, lag feature generation for time series data, categorical data encoding, text data vectorization, anomaly detection, and data normalization. For example, it analyzes past inventory trends, inventory levels, and incoming / outgoing quantities to identify situations such as "a sharp decline in beverage inventory when events coincided just before a public holiday and the weather deteriorated" or "a shortage of workers." Complex cases such as "There was a delay in shipment on [date]", "Waste losses increased during the week when many refrigerated products were nearing their expiration dates", and "There were many delivery delays on the day a delivery vehicle broke down" are extracted as features, and a prompt message such as "Product ID: A001, Period: Next week, Input data: Inbound / outbound data for the past 3 years, inventory data, regional demand index, weather data, traffic congestion level, event impact level, number of workers, delivery performance data, expiration date data, storage area information. Based on the above data, calculate the optimal inventory level for next week and the importance of demand fluctuation factors." is sent to the generative AI. The Generative AI Integration Department sends instructions and data from the Demand Forecasting Department and the Inventory Optimization Proposal Department to the Generative AI in an appropriate prompt format, receives the output results from the Generative AI, and returns them to each department. For example, it comprehensively assesses multiple factors and suggests the optimal inventory quantity, such as "The event impact is high and the demand index is high, so we should secure more inventory than usual," "The weather index is low, so there is a risk of delivery delays, and we need to have ample inventory," "There are enough workers, so there is ample capacity for receiving and shipping," or "There are many products nearing their expiration date, so we recommend ordering with the risk of waste in mind."The Inventory Optimization Proposal Department comprehensively considers multiple constraints, such as the demand forecast value for each product output from the Demand Forecasting Department, the current inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, worker utilization status, delivery vehicle utilization status, delivery plan, workload, storage area availability, inventory status of external warehouses, and the possibility of inventory transfer between multiple locations. Using generative AI, it automatically generates the optimal inventory quantity, inbound / outbound timing, delivery plan, workload distribution plan, and inventory transfer plan for each product. For example, "For product A, the demand forecast value for next week is 500 units, the current inventory is 100 units, the order lead time is 3 days, the minimum order unit is 50 units, the warehouse space limit is 600 units, and the logistics company's delivery frequency is 2 times a week." Under the condition that "there have been two stockouts in the last six months, calculate the order quantity that minimizes ordering and inventory costs," the system calculates the optimal order quantity and receiving timing. When the impact of events is high or the weather index is low, it takes into account the risk of delivery delays and sudden increases in demand, and proposes securing ample inventory and bringing forward delivery plans. If there is a shortage of workers, it takes into account the constraints of receiving and shipping processing capacity and automatically generates receiving and shipping schedules that level the workload. Simultaneously, it proposes inventory sharing between multiple locations and optimization of delivery routes according to the operating status of delivery vehicles and the congestion of delivery routes. For products requiring temperature control, such as refrigerated goods and pharmaceuticals, it also takes into account the temperature and humidity history of the storage area and the operating status of refrigerated vehicles, and proposes inventory management that minimizes quality risks. The field input unit is where on-site personnel at the logistics center input real-time information that occurs on-site, such as sudden large orders from delivery destinations, unexpected worker absences, accidents or equipment failures in the warehouse, the decision to hold new local events, notifications of delivery delays from logistics companies, urgent requests for inventory transfers from external warehouses, and temperature deviations due to refrigeration equipment failures, via terminals. This information is then immediately reflected in the server-side generation AI, allowing demand forecasts and inventory optimization suggestions to be recalculated in real time based on the latest on-site conditions.The user interface section displays operational suggestions on the screen for logistics center operators and on-site staff, including optimal inventory levels, inbound and outbound timing, delivery plans, workload distribution plans, inventory transfer plans, expiration date alerts, waste schedule lists, worker shift adjustment plans, and delivery vehicle operation plans. For example, it presents specific and practical suggestions from multiple perspectives, such as "The recommended inventory level for product A is 400 units, we recommend two inbound shipments this week, and plan additional deliveries before the event," "On Wednesdays when there is a shortage of workers, we will reduce inbound and outbound operations and concentrate them on Thursday," "On Fridays when traffic congestion is expected, we will change the delivery route and recommend early morning deliveries," "Since there are many refrigerated products nearing their expiration dates, we will conduct a sales promotion campaign this week to accelerate inventory consumption," and "We will transfer inventory from external warehouses on Monday to prepare for a surge in demand on Tuesday." Staff members use this information to make operational decisions and can continuously optimize the entire system by inputting additional on-site information as needed. The Notification & Alerts Department detects operational risks and anomalies in real time, such as approaching expiration dates, insufficient stock, excess stock, delivery delays, labor shortages, equipment failures, urgent requests from external warehouses, traffic disruptions, and sudden weather changes, and immediately notifies the staff's terminals and the administrator's email / messaging system. For example, it can detect situations such as "Stock of product B is expected to run out today," "Temperature deviation detected in refrigeration equipment," "Delivery delays due to delivery vehicle breakdown," "Shift readjustment required due to sudden worker absence," and "Emergency stock transfer request received from external warehouse," thereby enhancing on-site responsiveness. Specific examples of prompts to be fed into the generating AI include: "Product ID: A001, Period: Next week, Input data: Inbound / outbound data for the past 3 years, inventory data, regional demand index, weather data, traffic congestion level, event impact level, number of workers, delivery performance data, expiration date data, storage area information. Based on the above data, calculate the optimal inventory level and the importance of demand fluctuation factors for next week." and "For refrigerated product B, propose an operational plan to minimize the risk of waste based on the number of products nearing their expiration date this week, inventory level, delivery plan, storage area temperature history, worker shifts, and delivery vehicle operating status."Thus, the system according to this embodiment can optimize inventory management, inbound and outbound planning, delivery planning, workload distribution, quality control, risk management, and external collaboration in a logistics center in a multifaceted and real-time manner through the cooperation of each component. This enables the realization of an advanced operational support system that simultaneously achieves the prevention of stockouts and excess inventory, cost reduction, improved work efficiency, reduced quality risks, improved customer satisfaction, and enhanced on-site responsiveness.

[0074] (Implementation Steps)

[0075] Step 1: Inbound / Outbound and External Data Collection and Integration At the logistics center, the Inbound / Outbound Data Acquisition Department and the External Information Linkage Department work together to collect and store inventory data in real time, including the number of incoming and outgoing items, inventory quantity, incoming date, outgoing date, expiration date, disposal history, return history, inventory turnover rate, lot number, storage area, and temperature control information for all products. Furthermore, diverse external data such as demand indices for each delivery destination region, weather information (temperature, precipitation, snowfall, typhoons, heat waves, cold waves, sudden weather changes), traffic congestion (congestion status on major roads, construction / accident information, real-time location information of delivery vehicles), event impact (local festivals, sports competitions, concerts, exhibitions, opening / renovation of commercial facilities, school holidays, sales information, SNS buzz), number of workers, worker shifts, delivery performance data, delivery destination characteristics, vehicle operating status, worker skills and health status, logistics provider operating status, inventory status of external warehouses, and operating status of competing logistics centers are also automatically acquired and integrated, and centrally managed by the Database Management Department.

[0076] Step 2: Data preprocessing and feature generation The database management department performs preprocessing on multidimensional, multitemporal data accumulated by the department, including missing value imputation, outlier removal, feature engineering, lag feature generation for time series data, categorical data encoding, text data vectorization, anomaly detection, and data normalization. For example, complex cases such as "beverage inventory plummeted when events coincided just before a public holiday and the weather worsened," "shipping delays occurred on days when there was a shortage of workers," "waste losses increased during weeks when refrigerated products were nearing their expiration dates," and "delivery delays were frequent on days when delivery vehicles broke down" are extracted as features from past inbound, outbound, and inventory trends.

[0077] Step 3: Demand forecasting using generative AI The demand forecasting unit inputs pre-processed data into the generation AI via the generation AI integration unit, and accurately predicts the optimal inventory level and the importance of demand fluctuation factors for each product. Specific examples of prompts to be fed into the generation AI include: "Product ID: A001, Period: Next week, Input data: Inbound / outbound data for the past 3 years, inventory data, regional demand index, weather data, traffic congestion level, event impact level, number of workers, delivery performance data, expiration date data, storage area information. Based on the above data, calculate the optimal inventory level and the importance of demand fluctuation factors for next week." and "For refrigerated product B, propose an operational plan to minimize waste risk based on this week's number of products nearing their expiration date, inventory level, delivery plan, storage area temperature history, worker shifts, and delivery vehicle operating status."

[0078] Step 4: Inventory Optimization and Operational Proposal Generation The Inventory Optimization Proposal Department comprehensively considers multiple constraints, such as the demand forecast value for each product output from the Demand Forecasting Department, the current inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, worker utilization status, delivery vehicle utilization status, delivery plan, workload, storage area availability, inventory status of external warehouses, and the possibility of inventory transfer between multiple locations. Using a generative AI, it automatically generates the optimal inventory quantity, receiving and shipping timing, delivery plan, workload distribution plan, and inventory transfer plan for each product. For example, it sends a prompt to the generative AI such as, "For product A, the demand forecast value for next week is 500 units, the current inventory is 100 units, the order lead time is 3 days, the minimum order unit is 50 units, the warehouse space limit is 600 units, the logistics provider delivers twice a week, there has been a history of stockouts twice in the last six months, calculate the order quantity that minimizes ordering cost and inventory cost," and calculates the optimal order quantity, receiving timing, and delivery plan.

[0079] Step 5: Enter on-site information and reflect it in real time. The on-site input unit receives real-time information from logistics center staff via terminals, such as sudden large orders from delivery destinations, unexpected worker absences, warehouse accidents or equipment failures, decisions to hold new local events, delivery delay notifications from logistics companies, urgent inventory transfer requests from external warehouses, and temperature deviations due to refrigeration equipment failures. This information is immediately reflected in the server-side generation AI. As a result, demand forecasts and inventory optimization suggestions are recalculated in real time based on the latest on-site conditions.

[0080] Step 6: Present operational proposal and send notification alerts. The user interface unit displays operational proposals on the screen for logistics center operators and field staff, including optimal inventory levels, inbound and outbound timing, delivery plans, workload distribution proposals, inventory transfer proposals, expiration date alerts, disposal schedules, worker shift adjustment proposals, and delivery vehicle operation plans, all distributed from the server. The notification and alert unit detects operational risks and anomalies in real time, such as expiration dates, insufficient inventory, excess inventory, delivery delays, worker shortages, equipment failures, urgent requests from external warehouses, traffic disruptions, and sudden weather changes, and immediately notifies staff members' terminals and administrators' email and messaging systems.

[0081] (Specific use cases)

[0082] For example, let's consider the application of the present invention's system to operations in which a large-scale logistics center located in an urban area delivers a wide variety of products, such as beverages, daily necessities, refrigerated foods, and pharmaceuticals, to numerous retail stores, restaurants, and medical institutions. The inbound / outbound data acquisition unit collects in real time the number of incoming and outbound items, inventory quantity, inbound date, outbound date, expiration date, disposal history, return history, inventory turnover rate, lot number, storage area, temperature control information, etc., for each product, and stores them in the server's database management unit. The External Information Integration Department automatically acquires data such as demand indices for each delivery destination region, weather information (temperature, precipitation, typhoons, heat waves, cold waves, etc.), traffic congestion (congestion on major roads, construction / accident information, location information of delivery vehicles), event impact (local festivals, sports competitions, exhibitions, opening of commercial facilities, long holidays, sales information, trending topics on social media), number of workers, worker shifts, delivery performance data, delivery destination characteristics, vehicle operating status, worker skills and health status, logistics provider operating status, inventory status of external warehouses, and operating status of competing logistics centers, and centrally manages this data in the Database Management Department.

[0083] The demand forecasting unit preprocesses this multidimensional, multi-point-time data and inputs it into the generating AI via the generating AI collaboration unit. For example, for a major product A in the beverage category, it generates a prompt message to calculate the optimal inventory level and the importance of demand fluctuation factors for next week, based on the past three years of inbound / outbound data, inventory data, regional demand index, weather data, traffic congestion level, event impact level, number of workers, delivery performance data, expiration date data, and storage area information, and sends it to the generating AI. A specific example of a prompt would be: "Product ID: A001, Period: Next week, Input data: Inbound / outbound data, inventory data, regional demand index, weather data, traffic congestion level, event impact level, number of workers, delivery performance data, expiration date data, storage area information for the past three years. Based on the above data, calculate the optimal inventory level and the importance of demand fluctuation factors for next week." The generating AI comprehensively assesses multiple factors, such as "the event has a high impact and demand index, so we should secure more inventory than usual," "the weather index is low, so there is a risk of delivery delays and we need ample inventory," "we have enough workers, so we have sufficient capacity for receiving and shipping," and "we have many products nearing their expiration date, so we recommend ordering with the risk of waste in mind," and then suggests the optimal inventory quantity.

[0084] The Inventory Optimization Proposal Unit comprehensively considers multiple constraints, such as the demand forecast value for each product output from the Demand Forecasting Unit, the current inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, worker utilization status, delivery vehicle utilization status, delivery plan, workload, storage area availability, inventory status of external warehouses, and the possibility of inventory transfer between multiple locations. Using a generating AI, it automatically generates the optimal inventory quantity, receiving and shipping timing, delivery plan, workload distribution plan, and inventory transfer plan for each product. For example, it sends a prompt message to the generating AI such as, "For product A, the demand forecast value for next week is 500 units, the current inventory is 100 units, the order lead time is 3 days, the minimum order unit is 50 units, the warehouse space limit is 600 units, the logistics provider delivers twice a week, there have been 2 stockouts in the last 6 months, calculate the order quantity that minimizes ordering cost and inventory cost," and calculates the optimal order quantity, receiving timing, and delivery plan.

[0085] The on-site input unit allows logistics center staff to input real-time information from terminals that occurs on-site, such as sudden large orders from delivery destinations, unexpected worker absences, accidents or equipment failures in the warehouse, the decision to hold new local events, delivery delay notifications from logistics companies, urgent inventory transfer requests from external warehouses, and temperature deviations due to refrigeration equipment failures. This information is then immediately reflected in the server-side AI generation system. As a result, demand forecasts and inventory optimization suggestions are recalculated in real time based on the latest on-site conditions.

[0086] The user interface unit displays operational suggestions on the screen for logistics center operators and field staff, including optimal inventory levels, inbound and outbound timing, delivery plans, workload distribution proposals, inventory transfer proposals, expiration date alerts, disposal schedules, worker shift adjustment proposals, and delivery vehicle operation plans, all distributed from the server. The notification and alert unit detects operational risks and anomalies in real time, such as expiration dates, insufficient inventory, excess inventory, delivery delays, worker shortages, equipment failures, urgent requests from external warehouses, traffic disruptions, and sudden weather changes, and immediately notifies the staff's terminals and the administrator's email / messaging system.

[0087] Thus, the system of the present invention, in multi-product, multi-location distribution operations at large-scale urban logistics centers, integrates and analyzes vast and diverse data, automatically generates highly accurate demand forecasts and optimal inventory, distribution, and work plans using AI, and immediately reflects real-time information from the field. This simultaneously prevents stockouts and excess inventory, reduces costs, improves work efficiency, reduces quality risks, improves customer satisfaction, and enhances on-site responsiveness.

[0088] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0089] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0090] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0091] For example, the data collection unit is implemented by the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the smart device 14 and processes it with the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the output device 40 of the smart device 14 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the devices and control units is not limited to the examples described above and can be modified in various ways.

[0092] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0093] [Second Embodiment]

[0094] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0095] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0096] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0097] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and communication interface 44 are also connected to the bus 52.

[0098] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0099] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0100] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0101] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0102] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0103] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0104] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the identification processing unit 290 using these models.

[0105] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is realized by the data processing device 12 and the smart glasses 214. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0106] (Example 1)

[0107] Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.

[0108] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.

[0109] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0110] Furthermore, the processing by the data processing system 10 described above is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0111] For example, the data collection unit is implemented by the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the smart glasses 214 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the speaker 240 of the smart glasses 214 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0112] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0113] [Third Embodiment]

[0114] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0115] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0116] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0117] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, display 343, and communication interface 44 are also connected to the bus 52.

[0118] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0119] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0120] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0121] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0122] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0123] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0124] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0125] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is implemented by the data processing device 12 and the headset terminal 314. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0126] (Example 1) Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.

[0127] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.

[0128] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0129] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0130] For example, the data collection unit is implemented by the control unit 46A of the headset terminal 314 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the headset terminal 314 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the speaker 240 and display 343 of the headset terminal 314 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0131] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0132] [Fourth Embodiment]

[0133] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0134] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0135] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0136] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, controlled object 443, and communication interface 44 are also connected to the bus 52.

[0137] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0138] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0139] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0140] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0141] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0142] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0143] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0144] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0145] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is implemented by the data processing device 12 and the robot 414. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0146] (Example 1) Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.

[0147] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.

[0148] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0149] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or external devices, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or external devices.

[0150] For example, the data collection unit is implemented by the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the robot 414 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using a generation AI. For example, the serving unit is implemented by the speaker 240 and controlled object 443 of the robot 414 and the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0151] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0152] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0153] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0154] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0155] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0156] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0157] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0158] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0159] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0160] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0161] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0162] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0163] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0164] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0165] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0166] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0167] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0168] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0169] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0170] Furthermore, the following additional information is disclosed regarding the above explanation.

[0171] (Claim 1)

[0172] A logistics center operation support system comprising an inbound / outbound data acquisition unit, a demand forecasting unit, and an inventory optimization proposal unit, wherein the inbound / outbound data acquisition unit centrally collects and integrates various data on a server, such as past inbound / outbound quantities, inventory quantities, demand indices for each delivery destination region, weather information, traffic congestion, event impact, number of workers, delivery performance data, delivery destination characteristics, vehicle operating status, and worker shift information for each product, and the demand forecasting unit inputs the various data obtained from the inbound / outbound data acquisition unit into a generation system AI, and uses the ID of each product, inbound / outbound quantities, regional demand index, weather index, traffic congestion, event impact, and number of workers as explanatory variables to determine each product based on past performance. The system is characterized by its ability to accurately predict the optimal inventory level, calculate demand forecast values ​​to reduce the risk of stockouts and excess inventory, and then, using generative AI, automatically generate the optimal inventory level, inbound / outbound timing, and delivery plan for each product, taking into account the demand forecast values ​​calculated by the demand forecasting unit, as well as multiple constraints such as the current inventory status, order lead time, order unit, logistics constraints, inventory space, stockout risk, excess inventory risk, order cost, inventory cost, waste risk, worker utilization status, delivery vehicle utilization status, delivery plan, and workload, and outputting recommended inventory levels and optimal operational proposals to logistics center operators. (Claim 2)

[0173] The system according to claim 1, characterized in that, when training a demand forecasting model using generative AI, additional information such as sudden weather changes, local events, traffic disruptions, seasonal variations, promotional measures, characteristics of delivery destinations, past stockout / excess inventory history, changes in worker shifts, vehicle breakdowns or operational constraints are added as input features as external factors that affect demand fluctuations for each product, and the system improves demand forecasting accuracy by simultaneously considering these complex factors. (Claim 3)

[0174] The system according to claim 1, characterized in that the inventory optimization proposal department, when calculating inventory volume, inbound / outbound timing, and delivery plans based on demand forecasts, executes an algorithm using generative AI that simultaneously optimizes multiple constraints and cost factors such as logistics costs, worker workload, delivery vehicle operating efficiency, inventory space constraints, ordering cycle, delivery route optimization, emergency delivery risk, inventory sharing between multiple locations, and real-time information from on-site personnel, thereby automatically calculating the optimal inventory operation and delivery plan. [Explanation of symbols]

[0175] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. A system comprising: a data acquisition unit that accepts past sales data, sales data, and regional customer demographic data as input; a demand forecasting unit that integrates sales history, sales history, and regional attribute information for each store acquired from the data acquisition unit and predicts demand for each store and each product using generative AI; and an order proposal unit that calculates the optimal inventory order quantity for each store and each product based on the demand forecast values ​​calculated by the demand forecasting unit and outputs order proposal information to store staff or agency staff, wherein the demand forecasting unit inputs multiple attribute information such as store ID, product ID, month, past sales figures, past sales figures, regional age group, and regional income level as input features into a generative AI model and predicts future demand with high accuracy while considering past performance data and regional characteristics; and the order proposal unit calculates the optimal order quantity by comprehensively considering the demand forecast values, inventory status, order lead time, order unit, stockout risk, and excess inventory risk, and presents specific order proposals to staff via a user interface.

2. The system according to claim 1, characterized in that, when training a demand forecasting model using generative AI, additional information such as regional characteristics, seasonal fluctuations, promotional event information, past stockout history, and trends of competing stores, which differ for each store and each product, is incorporated as input features, and the accuracy of demand forecasting is improved by taking these complex factors into consideration.

3. The system according to claim 1, characterized in that the ordering proposal department, when calculating the order quantity based on demand forecast values, executes an algorithm using generative AI to simultaneously optimize multiple constraints and cost factors such as inventory costs, ordering costs, opportunity cost due to stockouts, waste risk due to excess inventory, logistics constraints, ordering cycles, inventory space constraints for each store, and the past ordering trends of the person in charge, and automatically calculates the optimal ordering proposal.

Citation Information

Patent Citations

  • Persona chatbot control method and system

    JP2022180282A