system

Abstract

Provide a system. 【Solution means】 Means for collecting sales information on each route and accumulating it in an information storage device, Means for constructing a model for demand prediction based on the information using a machine learning algorithm, Means for designing an inventory allocation plan based on the predicted demand, Means for presenting the allocation plan to a user using a display device, Means for collecting actual sales results as feedback and improving the model, When operating the inventory allocation plan, means for visually representing inventory liquidity via an information terminal and facilitating information confirmation by the user using voice support technology, Means for enhancing prediction accuracy by using prompt sentences through a generated AI model, A system including.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Inventory management in consumer sales agencies currently depends largely on manual work, which consumes a great deal of labor and time and is prone to overstocking or understocking. In addition, since inventory management that fully takes into account the characteristics of each distribution channel is not carried out, there is also a problem that efficient inventory allocation is difficult. In order to achieve efficient and accurate inventory management and improve the effectiveness of operations, automation and improvement of prediction accuracy are required.

Means for Solving the Problems

[0005] This invention provides a means for collecting sales data from each distribution channel and constructing a model for demand forecasting based on that data using a machine learning algorithm. Furthermore, it includes a means for formulating an inventory allocation plan based on the predicted demand and providing that plan to the user. By collecting actual sales performance as feedback, the accuracy of the machine learning model is continuously improved, enabling rapid and efficient inventory management that takes into account the characteristics of each distribution channel. In addition, by considering warehouse capacity limitations and transportation costs when formulating the inventory allocation plan, optimized inventory allocation becomes possible.

[0006] "Distribution channels" refer to the stages, intermediate points, and channels through which a product passes from the manufacturer to the consumer.

[0007] "Sales data" refers to records of product sales information, sales volume, sales period, and buyer characteristics for a specific period.

[0008] "Demand forecasting" is the process of predicting future fluctuations in demand for products and services based on past data.

[0009] A "model" is the structure of a mathematical method or algorithm used to analyze patterns in data.

[0010] A "machine learning algorithm" is a computational method that allows computers to learn from data and autonomously perform tasks such as prediction and classification.

[0011] An "inventory allocation plan" is a strategy and schedule for determining how to allocate available inventory through the distribution channels based on projected demand.

[0012] A "user" refers to an individual or organization that uses the system, specifically the person responsible for inventory management.

[0013] "Feedback" refers to actual sales performance and information provided by users of a system, and is used to improve the system.

[0014] An "optimization algorithm" is a computational method for obtaining the best possible result under specific constraints, and is typically used to solve maximization or minimization problems under constraints.

[0015] "Warehouse capacity limit" refers to the maximum number of goods that a warehouse can physically store.

[0016] "Transportation costs" refer to the total expenses required to move goods from one location to another. [Brief explanation of the drawing]

[0017] [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. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0019] First, the language used in the following description will be explained.

[0020] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0021] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0022] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] 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 A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] 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).

[0029] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] 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.

[0032] 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.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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 a 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 program 60 executed on the RAM 48.

[0037] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0038] This invention provides a system for achieving efficient and highly accurate inventory management, and a specific embodiment thereof is shown. The system mainly consists of three elements: a server, a terminal, and a user.

[0039] server

[0040] The server is responsible for collecting sales data from each distribution channel and storing it in a database. This allows for demand forecasting tailored to the characteristics of each sales channel, based on the data accumulated for each channel. The server uses machine learning algorithms to build models and perform highly accurate demand forecasts. Furthermore, the server develops an inventory allocation plan based on the predicted demand and provides that plan to the terminals.

[0041] terminal

[0042] The terminal visually presents the inventory allocation plan provided by the server to the user. Using visualization tools such as charts and graphs, the terminal helps the user intuitively understand inventory flow and demand forecasts. This allows the terminal to support the user in making quick decisions.

[0043] User

[0044] Users receive inventory information via their terminals and incorporate it into their actual on-site operations. Users also provide sales performance as feedback to the system, and the server uses this feedback to continuously improve the demand forecasting model. As a result, the system learns over time, enabling more accurate and faster inventory management.

[0045] Specific example

[0046] For example, if a sudden surge in demand for used devices is predicted at a particular store, the server immediately processes this information and instructs the user via the device on the optimal inventory quantity. Based on this instruction, the user secures the necessary inventory and provides it to the store. After the sale is completed, the user provides feedback on the performance data, and the server uses this data to improve the accuracy of predictions for the next cycle.

[0047] In this way, the system automates everything from sales data collection to feedback, reducing inventory surpluses and shortages, and enabling efficient inventory management tailored to the characteristics of each distribution channel.

[0048] The following describes the processing flow.

[0049] Step 1:

[0050] The server periodically collects sales data from each distribution channel. The collected data is stored in a database via queries, and missing or outlier values ​​are corrected through a data cleaning process.

[0051] Step 2:

[0052] The server applies machine learning algorithms to clean sales data to build a demand forecasting model. The model is optimized to take into account channel-specific patterns and seasonality, and is validated on a test dataset to evaluate its accuracy.

[0053] Step 3:

[0054] The server uses a trained demand forecasting model to predict demand for the next cycle. New data inputs are also incorporated into the demand forecast, and the forecast results are stored in a database.

[0055] Step 4:

[0056] The server develops an inventory allocation plan based on predicted demand. Here, optimization algorithms such as linear programming are used to calculate the optimal inventory allocation, taking into account constraints such as warehouse capacity limits and transportation costs.

[0057] Step 5:

[0058] The server sends inventory allocation plans and forecast data to the terminal. The terminal provides an interface to visually present information to the user, allowing them to intuitively understand inventory flow and demand forecasts.

[0059] Step 6:

[0060] Users check inventory information provided from their terminals and perform actual inventory management operations. They issue necessary inventory transfer and procurement instructions, and prepare for sales.

[0061] Step 7:

[0062] Users provide feedback on their actual sales performance to the server via their devices. The server uses this feedback data to readjust its demand forecasting model, enabling more accurate predictions in the next cycle.

[0063] (Example 1)

[0064] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0065] Inventory management systems are required to respond immediately to fluctuations in sales information, make more accurate demand forecasts, and achieve efficient inventory allocation. However, existing systems fail to adequately consider the characteristics of each distribution channel, which can lead to decreased forecast accuracy and inventory shortages or surpluses. Furthermore, there is a lack of established methods for effectively utilizing feedback information to improve models, making rapid decision-making difficult.

[0066] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0067] In this invention, the server includes means for collecting sales information along the distribution route and storing it in an information storage device, means for constructing a forecasting model for demand forecasting based on the information using machine learning techniques, and means for establishing an inventory allocation plan based on the forecasted demand. This makes it possible to improve forecasting accuracy by considering the characteristics of each distribution route and using a generative AI model, and to achieve optimized inventory allocation that takes into account the capacity limitations of storage facilities and logistics costs.

[0068] "Distribution channels" refer to the series of paths or channels involved in the process by which goods or services are delivered from producers to consumers.

[0069] "Sales information" refers to data related to the sales of goods and services, including purchase date and time, sales quantity, and customer information.

[0070] An "information storage device" refers to a storage medium or storage system that can store data and retrieve it as needed.

[0071] "Demand forecasting" refers to the analysis and methods used to predict the size and fluctuations of future demand.

[0072] A "predictive model" refers to a mathematical or algorithmic structure built to predict future events or trends based on data.

[0073] "Machine learning techniques" refer to algorithms and processes that learn patterns from data and use them to make future predictions and classifications.

[0074] An "inventory allocation plan" refers to a plan or strategy for optimally positioning and allocating product inventory to each distribution point based on demand forecasts.

[0075] A "generative AI model" refers to a model that uses artificial intelligence technology to learn from data and is generated to respond to new data or specific conditions.

[0076] A "storage facility" refers to a physical structure or location designed for the storage of goods or materials.

[0077] An "optimization method" refers to a mathematical or algorithmic technique for utilizing resources and conditions in the most efficient way to achieve a specific objective.

[0078] Modes for carrying out the invention

[0079] This invention provides an efficient and highly accurate inventory management system. The specific implementation of this system is described below.

[0080] server

[0081] The server's role is to collect sales information from each distribution channel and store it in a data storage device. Specifically, it retrieves data from multiple sales channels via APIs. This data includes product identification information, the number of units sold, and the sales date. The data is stored using a database management system (e.g., MySQL®). The server preprocesses and cleanses the data using Python or R, and then builds a demand forecasting model using machine learning techniques (e.g., TENSORFLOW® or Scikit-learn). Furthermore, the server develops an optimal inventory allocation plan based on this forecast data and provides it to the terminal.

[0082] terminal

[0083] The terminal functions to visually present inventory allocation plans provided by the server to the user. Visualization software such as Microsoft® Power BI and Tableau is used to ensure intuitive understanding for the user. Inventory flows and projected demand fluctuations can be displayed in graph and dashboard formats. This allows the terminal to help users make quick and effective decisions.

[0084] User

[0085] Users adjust and replenish inventory based on information provided through their terminals. Furthermore, users input sales performance data into the system as feedback. This includes sales figures, return information, and customer feedback. This feedback data is sent to a server and used to continuously improve the demand forecasting model.

[0086] Specific example

[0087] For example, if a sudden surge in demand for a new product is predicted, the server immediately processes the information and plans the optimal inventory allocation for each store. This information is notified to the user via a terminal, allowing the user to quickly secure the necessary inventory based on the instructions. After sales are complete, the user provides feedback on the performance data, which the server uses to improve future forecasts.

[0088] Example of a prompt

[0089] "Enter new sales data into the server and forecast inventory demand for the next month. Visualize the results, send them to the terminal, and improve the model based on user feedback."

[0090] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0091] Step 1:

[0092] The server collects sales information from each distribution channel and stores it in its data storage device. The input at this stage is raw data from the sales channels, specifically sales figures, dates, and product identification information obtained via APIs. This data is structured and stored using a database management system. The output is organized and stored in a format suitable for subsequent data analysis. Specifically, the server periodically calls APIs to retrieve and store new data.

[0093] Step 2:

[0094] The server preprocesses the stored data. The input is the sales information organized in Step 1. Data processing includes imputing missing values ​​and handling outliers, and is performed using Python's data analysis library. The output is a clean and consistent dataset, which is used to build the predictive model. Specifically, the server generates a data frame and performs the necessary data cleansing.

[0095] Step 3:

[0096] The server builds a demand forecasting model using pre-processed data. The input is a clean dataset, and the model is trained using a machine learning framework. By applying a generative AI model, future demand forecasts are made based on past sales trends. The output is the demand forecast result, which serves as the basis for inventory allocation planning. Specifically, the server optimizes the model's hyperparameters and executes the forecasting algorithm.

[0097] Step 4:

[0098] The server develops an inventory allocation plan based on predicted demand. The input is the demand forecast, and an optimization algorithm is used to calculate the optimal inventory allocation to each sales channel. Factors to consider include distribution costs and warehouse capacity. The output is a specific inventory allocation plan for each sales channel. In practice, the server determines the optimal inventory allocation using linear programming and other optimization techniques.

[0099] Step 5:

[0100] The terminal visually displays the inventory allocation plan received from the server. The input is the inventory allocation plan provided by the server, and the terminal provides information to the user using a visualization tool. The output is intuitive visual information presented in a dashboard format. Specifically, the terminal generates graphs and charts and organizes and displays the data in a way that is easy for the user to understand.

[0101] Step 6:

[0102] The user adjusts inventory based on terminal information and inputs sales performance data as feedback. Input data includes sales volume, return information, and other performance data. The output is improved feedback data, contributing to improved future forecast accuracy. Specifically, the user supports the system's learning process by operating the terminal and accurately inputting feedback information.

[0103] (Application Example 1)

[0104] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0105] In logistics operations, improving the efficiency of inventory management and the accuracy of demand forecasting are crucial challenges. Traditional systems often fail to collect sales data quickly and accurately, leading to inventory shortages and surpluses, as well as increased costs. Furthermore, it is difficult for on-site users to easily understand inventory information and respond promptly.

[0106] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0107] In this invention, the server includes means for collecting sales information and storing it in an information storage device, means for constructing a demand forecasting model using a machine learning algorithm, and means for designing an inventory allocation plan. This enables rapid collection and analysis of sales information, and efficient inventory management through highly accurate demand forecasting. Furthermore, by visually representing inventory liquidity through an information terminal and using voice support technology, users can intuitively understand the information and make quick decisions.

[0108] "Route" is a term that refers to the points or paths that goods or information take to circulate.

[0109] An "information storage device" is an electronic device or system for accumulating various types of collected data and storing them in a format that makes them easily accessible when needed.

[0110] A "machine learning algorithm" is a method or computational procedure that allows a computer to learn patterns from data and perform predictions and classifications.

[0111] An "inventory allocation plan" is a strategic plan for appropriately allocating inventory of goods and materials to each location based on demand forecasts.

[0112] A "display device" is an electronic device that visually displays information and assists users in confirming and managing it.

[0113] "Users" refer to individuals who operate systems and devices, and who utilize information to perform tasks and make decisions.

[0114] "Voice assistance technology" is a technology that uses voice to give instructions and obtain information, enabling users to access information hands-free.

[0115] A "generative AI model" is an artificial intelligence model that learns from given data and prompts and is generated to address new problems.

[0116] A "prompt statement" refers to a document or command used when inputting instructions or data into an AI model.

[0117] The system that implements this application consists of three elements: a server, a terminal, and a user. Its aim is to create an efficient environment for inventory management.

[0118] The server collects sales information and stores it in an information storage device. The collected data is used to build a demand forecasting model using machine learning algorithms. Specifically, Flask is used for data communication, and TensorFlow is used to train the forecasting model. Based on the built model, the server designs an inventory allocation plan, which is then sent to terminals via information terminals.

[0119] The terminal is responsible for visually presenting the inventory allocation plan received from the server to the user. The terminal has an application built using React Native installed, enabling data visualization using Charts.js. Furthermore, it utilizes Google® Cloud Speech-to-Text as a voice assistance technology, providing users with the ability to access information hands-free.

[0120] Users use their devices to check and manage inventory information and allocation plans. The decisions made by users are sent to the server as feedback. This feedback is used to improve the model in the next cycle, enhancing the accuracy of the generated AI model.

[0121] To give a specific example, if a sudden fluctuation in demand for a particular product is predicted at a logistics center, the server immediately processes that information and instructs the terminal to allocate inventory optimally. Based on these instructions, the user optimizes inventory, enabling efficient logistics without surpluses or shortages.

[0122] An example of a prompt message for a generative AI model might be: "Create a program that forecasts demand for a specific product category and displays the results in a smartphone app."

[0123] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0124] Step 1:

[0125] The server collects sales information from each distribution channel and stores it in the information storage device. In this step, the server collects sales information into a database and organizes and stores the data. The input is raw sales data from each distribution channel, and the output is organized sales data stored in the information storage device. Data normalization and duplicate removal are performed.

[0126] Step 2:

[0127] The server uses accumulated sales data to execute a machine learning algorithm and build a demand forecasting model. In this step, the model is trained using tools such as TensorFlow. The input is organized sales data stored in an information storage device, and the output is a trained model specifically designed for demand forecasting. After data preprocessing, the model performs pattern recognition and generates forecast data.

[0128] Step 3:

[0129] The server predicts the demand for each product based on the built model and develops an inventory allocation plan. Here, the server performs predictions using the trained model and calculates the optimal inventory allocation based on the results. The input is the trained demand forecasting model and newly collected sales data, and the output is the inventory allocation plan. The forecasting algorithm identifies the supply-demand gap and automatically generates the allocation plan.

[0130] Step 4:

[0131] The terminal visually presents the inventory allocation plan received from the server to the user. In this step, the terminal uses Charts.js visualization via a React Native application. The input is the inventory allocation plan sent from the server, and the output is information presented visually and audibly on the screen and through voice assistance technologies. The terminal uses a voice assistant to inform the user of important information.

[0132] Step 5:

[0133] Users review inventory allocation plans via a terminal and take actions based on on-site conditions. They also send feedback to the server to improve the accuracy of future forecasts. In this step, users perform physical inventory operations based on the plan and provide feedback on the results. Inputs are the inventory allocation plan and on-site information displayed on the terminal, and outputs are the results of the actions taken and the feedback thereto. Based on the results, inventory optimization and model accuracy improvements are performed.

[0134] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0135] This invention combines an emotion engine with an inventory management system to optimize inventory allocation while considering the user's emotional state. The system is composed of three elements: a server, a terminal, and a user.

[0136] server

[0137] The server builds a demand forecasting model using machine learning algorithms based on sales data collected from various distribution channels. In addition, the emotion engine processes emotional data obtained from user input and actions and incorporates it into the demand forecasting model. The emotion engine analyzes user text input and voice data using natural language processing and sentiment analysis algorithms.

[0138] terminal

[0139] When the terminal presents the user with an inventory allocation plan, the emotion engine displays information tailored to the user's emotional state. For example, if the user is stressed, the terminal provides concise, real-time information and applies an interface to assist in decision-making.

[0140] User

[0141] Users can view inventory data and demand forecasts through their devices to make decisions for the next steps. Sentimental data obtained from user interactions is fed back to the server, which uses this data to continuously improve its forecasting models and displayed content.

[0142] Specific example

[0143] For example, if the emotion engine detects signs such as a user spending a long time working on the inventory management screen and their input speed slowing down, the server records this as a record, and the next time a similar situation is detected, the terminal will send a simplified notification. Also, if the emotion engine detects a state of excitement when demand for a particular product suddenly increases, the server will immediately recommend high-priority inventory management actions.

[0144] In this way, the system provides a form that supports efficient decision-making for users by performing highly accurate inventory management through the collection of sales data and the analysis of sentiment.

[0145] The following describes the processing flow.

[0146] Step 1:

[0147] The server periodically collects sales data from each distribution channel and stores it in a database. This includes sales history, inventory status, and seasonal sales characteristics. The data is then processed using data cleaning algorithms to make it ready for analysis.

[0148] Step 2:

[0149] The server uses an emotion engine to analyze the user's emotional state from their operation logs and input. The emotion engine uses natural language processing and machine learning models to analyze text and voice input and recognize the user's emotional state.

[0150] Step 3:

[0151] The server combines sales data and user sentiment data to build a demand forecasting model. This model is based on machine learning algorithms and is designed to maximize forecasting accuracy by taking into account the characteristics of each distribution channel.

[0152] Step 4:

[0153] The server develops an inventory allocation plan based on predicted demand. This may involve prioritization based on sentiment data, and optimization algorithms are used to ensure that all allocations are reasonable and meet sales criteria.

[0154] Step 5:

[0155] The terminal visually presents the inventory allocation plan received from the server to the user. It incorporates the results of the emotion engine to provide information that reassures the user. For example, if the user is experiencing high stress levels, the notification will offer an intuitive and simple action.

[0156] Step 6:

[0157] Users use their devices to review the proposed inventory allocation plan and make adjustments as needed. Users also send feedback from their devices to the server, providing information, particularly regarding the effectiveness of sentiment-based responses.

[0158] Step 7:

[0159] The server analyzes user feedback and actual sales results, and based on this, continuously improves its demand forecasting model and sentiment analysis algorithm. As a result, the system's accuracy and efficiency improve over time.

[0160] (Example 2)

[0161] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0162] Conventional inventory management systems rely solely on historical sales data for demand forecasting, resulting in low accuracy in predicting actual demand fluctuations. Furthermore, they fail to consider the impact of user emotional states on inventory allocation planning decisions, potentially leading to inefficient decision-making. Therefore, a more accurate inventory allocation method that reflects user emotional states is needed.

[0163] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0164] In this invention, the server includes means for collecting sales information from each distribution channel and storing it in an information storage device, means for constructing a structure for making demand forecasts based on the information using machine learning technology, and means for processing user emotional information using an emotional analysis engine and reflecting it in the demand forecasting structure. This enables highly accurate demand forecasting and inventory allocation that takes into account the emotional state of the user.

[0165] "Distribution channels" refer to the routes and networks through which products and services are delivered from producers to consumers.

[0166] "Sales information" refers to information that includes data on product sales and sales status, and is used for forecasting future demand.

[0167] An "information storage device" includes hardware and software for storing collected data and accessing and processing it as needed.

[0168] "Machine learning techniques" is a general term for algorithms and methods that can learn patterns from large amounts of data and perform predictions and classifications.

[0169] "Structure" refers to algorithms and data flows used to process information and build models, and is a crucial element in system design.

[0170] A "sentiment analysis engine" is a software technology that uses natural language processing and speech processing to analyze human emotions and utilize them as data.

[0171] "Users" refer to individuals who operate the system, input data, and make decisions.

[0172] "Emotional information" refers to data that indicates the emotional state of a user, analyzed based on their actions and inputs.

[0173] "Demand forecasting" refers to the process of predicting future demand for a product based on past sales information and current data.

[0174] This invention is a new type of system that combines an inventory management system and an emotion engine, and its core is based on data exchange between a server, a terminal, and a user.

[0175] server

[0176] The server primarily collects sales information from various distribution channels and stores it in a database, which serves as the information storage device. This process involves periodically executing data collection scripts and retrieving sales information via APIs and other data acquisition methods. The collected information is used to build demand forecasting models using machine learning techniques. Specifically, random forest and linear regression models are trained using Python's Scikit-learn. Furthermore, the server utilizes a sentiment analysis engine to analyze text input and audio data from users and extract sentiment information. Natural language processing libraries such as NLTK and Transformers are used for this process. The analyzed sentiment information is then incorporated as new features into the demand forecasting model, resulting in more accurate predictions.

[0177] terminal

[0178] When the terminal presents the inventory allocation plan transmitted from the server to the user, the information display is adjusted based on the results of the sentiment analysis engine. For example, if the user's stress level is detected, the information is provided in a simplified form in real time, and an interface is applied to support necessary decision-making. This makes it easier for the user to understand and make quick decisions.

[0179] User

[0180] Users access inventory data and demand forecasts through their terminals and make decisions for the next steps based on that information. Sentimental information and feedback data obtained through user interaction are fed back to the server and used for continuous improvement of the demand forecasting model and displayed content. In this way, the entire system forms a dynamic improvement cycle that takes user emotions into account, enhancing the user experience.

[0181] Specific example

[0182] For example, if the emotion engine detects a user's level of interest by frequently checking the inventory status of a particular product, it will prioritize displaying inventory data for that product the next time the user checks. In addition, if the analysis reveals that the user is experiencing fatigue due to slow input speed, the device can simplify the information and display only the most important data.

[0183] Examples of prompts for generative AI models

[0184] "Based on this week's sales data and user sentiment trends, please propose an inventory allocation plan for next week."

[0185] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0186] Step 1:

[0187] The server collects sales information from each distribution channel. The input is raw sales data obtained via APIs and other data acquisition methods. The server organizes this data and stores it in an information storage device. Data cleaning is performed to remove inaccurate data before it is stored in the database. The output is a clean set of sales information.

[0188] Step 2:

[0189] The server builds a demand forecasting model using sales information. The input is the set of sales information obtained in the previous step. The server uses Scikit-learn, a machine learning technique, to train the model by applying algorithms such as random forest and linear regression. During this process, distribution and trend analysis is performed, and a model for predicting future demand is output.

[0190] Step 3:

[0191] The server performs user sentiment analysis. The input is text or voice data from the user. The server analyzes this data using a natural language processing library and extracts the user's sentiment information. The sentiment analysis engine identifies emotional states such as positive, negative, or neutral, and provides sentiment information as output.

[0192] Step 4:

[0193] The server incorporates the acquired sentiment information into the demand forecasting model. The inputs are the demand forecasting model and the sentiment information. The server adds the sentiment information as a feature of the model and readjusts it. This readjustment enables highly accurate forecasts that take emotional states into account. The output is the demand forecasting model with sentiment reflected.

[0194] Step 5:

[0195] The terminal develops and presents an inventory allocation plan to the user based on a demand forecasting model that incorporates sentiment from the server. The inputs are the demand forecasting model and inventory data. The terminal analyzes the information and provides the user with an interface for visualization. During this process, the displayed content is adjusted according to the user's emotional state, and the adjusted inventory allocation plan is presented to the user as output.

[0196] Step 6:

[0197] The user reviews the presented inventory allocation plan and makes adjustments as needed. The input is the inventory allocation plan presented by the terminal. User interaction through the interface generates feedback information regarding the plan. The output is the information sent to the server as user feedback data.

[0198] Step 7:

[0199] The server improves the model based on user feedback. The inputs are feedback data and the current demand forecasting model. The server analyzes the feedback and tunes the model parameters as needed. This improves forecasting accuracy, resulting in an improved demand forecasting model as output.

[0200] (Application Example 2)

[0201] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0202] Traditional inventory management systems perform demand forecasting and inventory allocation based on sales data, but they do not take into account the emotional state of users, and therefore cannot be said to have fully achieved optimization of inventory allocation. In particular, when users are experiencing stress or anxiety, the way inventory information is provided may not be appropriate. In such situations, the efficiency of decision-making decreases, and as a result, the accuracy of inventory management is compromised.

[0203] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0204] In this invention, the server includes means for collecting sales data from each distribution channel and storing it in a database, means for constructing a demand forecasting model using a machine learning algorithm based on the data, and means for influencing the forecasted demand based on the user's emotional state. This makes it possible to optimize inventory allocation while taking into account the user's emotional state.

[0205] "Sales data" refers to records of how products were sold through each distribution channel.

[0206] A "database" is a collection of data built to efficiently store, manage, and retrieve collected information.

[0207] A "machine learning algorithm" is a computational method used to build predictive models that improve themselves using collected data.

[0208] "User emotional state" refers to information about the psychological state or changes a user exhibits when operating a system.

[0209] An "inventory allocation plan" is a strategy for efficiently distributing goods to each location based on predicted demand.

[0210] A "display device" is a device used to visually present information to a user.

[0211] "Feedback" is the process of returning actual performance data to the system and using it to improve models and plans.

[0212] The system that implements this application example includes a series of processes to optimize inventory management by taking into account the user's emotional state. A detailed explanation of how the system works is provided below.

[0213] The server collects sales data from each distribution channel and stores it in a database. This prepares the foundational data for demand forecasting. As a machine learning algorithm, "scikit-learn" is used to analyze the sales data and build a demand forecasting model. Furthermore, the sentiment engine extracts sentiment data from user input and actions, and analyzes it using natural language processing libraries such as "NLTK" and "TextBlob". This sentiment data is then reflected in the demand forecasting model.

[0214] The device displays information tailored to the user's emotional state, and if the user is feeling stressed, it presents information in a concise and concise manner. This allows the user to make quick and accurate decisions.

[0215] Based on the inventory data and forecast information displayed on the device, users decide what action to take next. Sentiment data obtained from user interactions is constantly fed back to the server and used to improve the accuracy of the model and optimize the displayed content.

[0216] A concrete example would be a scenario where an employee working at a logistics center gives a voice command via their smartphone, saying, "Tell me the latest demand forecast." In this case, if the system senses anxiety from the voice, it will concisely display only the most important information visually. An example of a prompt message would be, "Infer anxiety from the voice and display inventory information accordingly."

[0217] This system streamlines the inventory management process at logistics centers and reduces the psychological burden on users.

[0218] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0219] Step 1:

[0220] The server collects sales data from each distribution channel and stores it in a database. This sales data contains historical information about when and through which channels products were sold. Sales logs from each store and distributor are provided as input, and this data is organized and stored in the database as output. Specifically, the server performs data insertion into the database and consistency checks.

[0221] Step 2:

[0222] The server uses collected sales data to build a demand forecasting model based on a machine learning algorithm. Sales performance and distribution channel characteristics are provided as input data, and the demand forecasting model is generated as output. This process utilizes "scikit-learn" for data preprocessing, algorithm selection, model training, and evaluation of prediction accuracy.

[0223] Step 3:

[0224] The server analyzes voice or text data obtained from user input and actions using natural language processing tools such as "NLTK" and "TextBlob" to evaluate the user's emotional state. Voice data or text chat is provided as input, and the user's emotional state is obtained as output. Specific operations include text conversion, sentiment analysis, and identification of emotional tendencies.

[0225] Step 4:

[0226] The server updates the demand forecasting model based on the sentiment data obtained from the sentiment engine. Sentiment data is input, and a new demand forecast reflecting the sentiment is created as output. In this process, the model is retrained to incorporate the sentiment data.

[0227] Step 5:

[0228] The terminal receives the inventory allocation plan provided by the server and displays it to the user. Here, the information is simplified according to the user's emotional state. Inventory allocation information is received as input, and a visual display tailored to the user's emotions is produced as output. Specifically, filtering and highlighting of the displayed content are performed.

[0229] Step 6:

[0230] The user makes decisions based on the presented inventory allocation plan and feeds the results back to the server. The inventory allocation plan is displayed as input, and feedback data is generated as output. Specifically, the actions decided by the user are sent to the server and used for future model improvements of the system.

[0231] The specific processing unit 290 transmits the result of the specific processing 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.

[0232] Data generation model 58 is a 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> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. 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. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0233] 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.

[0234] [Second Embodiment]

[0235] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0236] 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.

[0237] 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).

[0238] 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, and camera 42 are also connected to the bus 52.

[0239] 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.

[0240] 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).

[0241] 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.

[0242] 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.

[0243] 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.

[0244] 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.

[0245] 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.

[0246] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0247] This invention provides a system for achieving efficient and highly accurate inventory management, and a specific embodiment thereof is shown. The system mainly consists of three elements: a server, a terminal, and a user.

[0248] server

[0249] The server is responsible for collecting sales data from each distribution channel and storing it in a database. This allows for demand forecasting tailored to the characteristics of each sales channel, based on the data accumulated for each channel. The server uses machine learning algorithms to build models and perform highly accurate demand forecasts. Furthermore, the server develops an inventory allocation plan based on the predicted demand and provides that plan to the terminals.

[0250] terminal

[0251] The terminal visually presents the inventory allocation plan provided by the server to the user. Using visualization tools such as charts and graphs, the terminal helps the user intuitively understand inventory flow and demand forecasts. This allows the terminal to support the user in making quick decisions.

[0252] User

[0253] Users receive inventory information via their terminals and incorporate it into their actual on-site operations. Users also provide sales performance as feedback to the system, and the server uses this feedback to continuously improve the demand forecasting model. As a result, the system learns over time, enabling more accurate and faster inventory management.

[0254] Specific example

[0255] For example, if a sudden surge in demand for used devices is predicted at a particular store, the server immediately processes this information and instructs the user via the device on the optimal inventory quantity. Based on this instruction, the user secures the necessary inventory and provides it to the store. After the sale is completed, the user provides feedback on the performance data, and the server uses this data to improve the accuracy of predictions for the next cycle.

[0256] In this way, the system automates everything from sales data collection to feedback, reducing inventory surpluses and shortages, and enabling efficient inventory management tailored to the characteristics of each distribution channel.

[0257] The following describes the processing flow.

[0258] Step 1:

[0259] The server periodically collects sales data from each distribution channel. The collected data is stored in a database via queries, and missing or outlier values ​​are corrected through a data cleaning process.

[0260] Step 2:

[0261] The server applies machine learning algorithms to clean sales data to build a demand forecasting model. The model is optimized to take into account channel-specific patterns and seasonality, and is validated on a test dataset to evaluate its accuracy.

[0262] Step 3:

[0263] The server uses a trained demand forecasting model to predict demand for the next cycle. New data inputs are also incorporated into the demand forecast, and the forecast results are stored in a database.

[0264] Step 4:

[0265] The server develops an inventory allocation plan based on predicted demand. Here, optimization algorithms such as linear programming are used to calculate the optimal inventory allocation, taking into account constraints such as warehouse capacity limits and transportation costs.

[0266] Step 5:

[0267] The server sends inventory allocation plans and forecast data to the terminal. The terminal provides an interface to visually present information to the user, allowing them to intuitively understand inventory flow and demand forecasts.

[0268] Step 6:

[0269] Users check inventory information provided from their terminals and perform actual inventory management operations. They issue necessary inventory transfer and procurement instructions, and prepare for sales.

[0270] Step 7:

[0271] Users provide feedback on their actual sales performance to the server via their devices. The server uses this feedback data to readjust its demand forecasting model, enabling more accurate predictions in the next cycle.

[0272] (Example 1)

[0273] Next, we will describe Example 1. 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".

[0274] Inventory management systems are required to respond immediately to fluctuations in sales information, make more accurate demand forecasts, and achieve efficient inventory allocation. However, existing systems fail to adequately consider the characteristics of each distribution channel, which can lead to decreased forecast accuracy and inventory shortages or surpluses. Furthermore, there is a lack of established methods for effectively utilizing feedback information to improve models, making rapid decision-making difficult.

[0275] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0276] In this invention, the server includes means for collecting sales information along the distribution route and storing it in an information storage device, means for constructing a forecasting model for demand forecasting based on the information using machine learning techniques, and means for establishing an inventory allocation plan based on the forecasted demand. This makes it possible to improve forecasting accuracy by considering the characteristics of each distribution route and using a generative AI model, and to achieve optimized inventory allocation that takes into account the capacity limitations of storage facilities and logistics costs.

[0277] The "distribution channel" refers to a series of paths or channels in the process by which goods and services are delivered from producers to consumers.

[0278] "Sales information" refers to data related to the sales of goods and services, including purchase date and time, sales quantity, customer information, etc.

[0279] An "information storage device" refers to a storage medium or storage system that can store data and retrieve it as needed.

[0280] "Demand forecasting" refers to the analysis and methods for predicting the magnitude and fluctuations of future demand.

[0281] A "prediction model" refers to a mathematical or algorithmic structure constructed to predict future events and trends based on data.

[0282] "Machine learning techniques" refer to algorithms and processes that learn patterns from data and perform future predictions and classifications.

[0283] An "inventory allocation plan" refers to a plan or strategy for optimally arranging and distributing the inventory of goods to each distribution point based on demand forecasting.

[0284] A "generative AI model" refers to a model generated using artificial intelligence technology to learn from data and respond to new data or specific conditions.

[0285] A "storage facility" refers to a physical structure or location designed to store goods and materials.

[0286] "Optimization techniques" refer to mathematical or algorithmic technologies for most efficiently utilizing resources and conditions to achieve a specific goal.

[0287] Modes for Implementing the Invention

[0288] This invention provides an efficient and highly accurate inventory management system. The specific implementation of this system is described below.

[0289] server

[0290] The server's role is to collect sales information from each distribution channel and store it in a data storage device. Specifically, it retrieves data from multiple sales channels via APIs. This data includes product identification information, the number of units sold, and the sales date. The data is stored using a database management system (e.g., MySQL). The server preprocesses and cleanses the data using Python or R, and then builds a demand forecasting model using machine learning techniques (e.g., TensorFlow or Scikit-learn). Furthermore, the server develops an optimal inventory allocation plan based on this forecast data and provides it to the terminal.

[0291] terminal

[0292] The terminal functions to visually present inventory allocation plans provided by the server to the user. Visualization software such as Microsoft Power BI and Tableau is used to ensure intuitive understanding for the user. Inventory flows and projected demand fluctuations can be displayed in graph and dashboard formats. This allows the terminal to help users make quick and effective decisions.

[0293] User

[0294] Users adjust and replenish inventory based on information provided through their terminals. Furthermore, users input sales performance data into the system as feedback. This includes sales figures, return information, and customer feedback. This feedback data is sent to a server and used to continuously improve the demand forecasting model.

[0295] Specific example

[0296] For example, if a sudden surge in demand for a new product is predicted, the server immediately processes the information and plans the optimal inventory allocation for each store. This information is notified to the user via a terminal, allowing the user to quickly secure the necessary inventory based on the instructions. After sales are complete, the user provides feedback on the performance data, which the server uses to improve future forecasts.

[0297] Example of a prompt

[0298] "Enter new sales data into the server and forecast inventory demand for the next month. Visualize the results, send them to the terminal, and improve the model based on user feedback."

[0299] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0300] Step 1:

[0301] The server collects sales information from each distribution channel and stores it in its data storage device. The input at this stage is raw data from the sales channels, specifically sales figures, dates, and product identification information obtained via APIs. This data is structured and stored using a database management system. The output is organized and stored in a format suitable for subsequent data analysis. Specifically, the server periodically calls APIs to retrieve and store new data.

[0302] Step 2:

[0303] The server preprocesses the stored data. The input is the sales information organized in Step 1. Data processing includes imputing missing values ​​and handling outliers, and is performed using Python's data analysis library. The output is a clean and consistent dataset, which is used to build the predictive model. Specifically, the server generates a data frame and performs the necessary data cleansing.

[0304] Step 3:

[0305] The server constructs a demand prediction model using the pre - processed data. The input is a clean dataset, and the model is trained using a machine learning framework. By applying the generative AI model, future demand prediction based on past sales trends is performed. The output is the demand prediction result, which serves as the basic information for the inventory allocation plan. As specific operations, the server optimizes the hyperparameters of the model and executes the prediction algorithm.

[0306] Step 4:

[0307] The server formulates an inventory allocation plan based on the predicted demand. The input is the demand prediction result, and an optimization algorithm is used to calculate the optimal allocation of inventory to each sales channel. Factors to be considered include distribution costs and warehouse capacity. The output is a specific inventory allocation plan for each sales channel. As specific operations, using linear programming or other optimization methods, the server determines the optimal allocation of inventory.

[0308] Step 5:

[0309] The terminal visually displays the inventory allocation plan received from the server. The input is the inventory allocation plan provided by the server, and a visualization tool is used to provide information to the user. The output is intuitive visual information presented in a dashboard format. As specific operations, the terminal generates graphs and charts, organizes and displays the data for easy viewing by the user.

[0310] Step 6:

[0311] The user adjusts inventory based on terminal information and inputs sales performance data as feedback. Input data includes sales volume, return information, and other performance data. The output is improved feedback data, contributing to improved future forecast accuracy. Specifically, the user supports the system's learning process by operating the terminal and accurately inputting feedback information.

[0312] (Application Example 1)

[0313] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0314] In logistics operations, improving the efficiency of inventory management and the accuracy of demand forecasting are crucial challenges. Traditional systems often fail to collect sales data quickly and accurately, leading to inventory shortages and surpluses, as well as increased costs. Furthermore, it is difficult for on-site users to easily understand inventory information and respond promptly.

[0315] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0316] In this invention, the server includes means for collecting sales information and storing it in an information storage device, means for constructing a demand forecasting model using a machine learning algorithm, and means for designing an inventory allocation plan. This enables rapid collection and analysis of sales information, and efficient inventory management through highly accurate demand forecasting. Furthermore, by visually representing inventory liquidity through an information terminal and using voice support technology, users can intuitively understand the information and make quick decisions.

[0317] "Route" is a term that refers to the points or paths that goods or information take to circulate.

[0318] An "information storage device" is an electronic device or system for accumulating various types of collected data and storing them in a format that makes them easily accessible when needed.

[0319] A "machine learning algorithm" is a method or computational procedure that allows a computer to learn patterns from data and perform predictions and classifications.

[0320] An "inventory allocation plan" is a strategic plan for appropriately allocating inventory of goods and materials to each location based on demand forecasts.

[0321] A "display device" is an electronic device that visually displays information and assists users in confirming and managing it.

[0322] "Users" refer to individuals who operate systems and devices, and who utilize information to perform tasks and make decisions.

[0323] "Voice assistance technology" is a technology that uses voice to give instructions and obtain information, enabling users to access information hands-free.

[0324] A "generative AI model" is an artificial intelligence model that learns from given data and prompts and is generated to address new problems.

[0325] A "prompt statement" refers to a document or command used when inputting instructions or data into an AI model.

[0326] The system that implements this application consists of three elements: a server, a terminal, and a user. Its aim is to create an efficient environment for inventory management.

[0327] The server collects sales information and stores it in an information storage device. The collected data is used to build a demand forecasting model using machine learning algorithms. Specifically, Flask is used for data communication, and TensorFlow is used to train the forecasting model. Based on the built model, the server designs an inventory allocation plan, which is then sent to terminals via information terminals.

[0328] The terminal is responsible for visually presenting the inventory allocation plan received from the server to the user. The terminal has an application built using React Native installed, enabling data visualization using Charts.js. Furthermore, it utilizes Google Cloud Speech-to-Text as a voice assistance technology, providing users with the ability to access information hands-free.

[0329] Users use their devices to check and manage inventory information and allocation plans. The decisions made by users are sent to the server as feedback. This feedback is used to improve the model in the next cycle, enhancing the accuracy of the generated AI model.

[0330] To give a specific example, if a sudden fluctuation in demand for a particular product is predicted at a logistics center, the server immediately processes that information and instructs the terminal to allocate inventory optimally. Based on these instructions, the user optimizes inventory, enabling efficient logistics without surpluses or shortages.

[0331] An example of a prompt message for a generative AI model might be: "Create a program that forecasts demand for a specific product category and displays the results in a smartphone app."

[0332] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0333] Step 1:

[0334] The server collects sales information from each distribution channel and stores it in the information storage device. In this step, the server collects sales information into a database and organizes and stores the data. The input is raw sales data from each distribution channel, and the output is organized sales data stored in the information storage device. Data normalization and duplicate removal are performed.

[0335] Step 2:

[0336] The server uses accumulated sales data to execute a machine learning algorithm and build a demand forecasting model. In this step, the model is trained using tools such as TensorFlow. The input is organized sales data stored in an information storage device, and the output is a trained model specifically designed for demand forecasting. After data preprocessing, the model performs pattern recognition and generates forecast data.

[0337] Step 3:

[0338] The server predicts the demand for each product based on the built model and develops an inventory allocation plan. Here, the server performs predictions using the trained model and calculates the optimal inventory allocation based on the results. The input is the trained demand forecasting model and newly collected sales data, and the output is the inventory allocation plan. The forecasting algorithm identifies the supply-demand gap and automatically generates the allocation plan.

[0339] Step 4:

[0340] The terminal visually presents the inventory allocation plan received from the server to the user. In this step, the terminal uses Charts.js visualization via a React Native application. The input is the inventory allocation plan sent from the server, and the output is information presented visually and audibly on the screen and through voice assistance technologies. The terminal uses a voice assistant to inform the user of important information.

[0341] Step 5:

[0342] Users review inventory allocation plans via a terminal and take actions based on on-site conditions. They also send feedback to the server to improve the accuracy of future forecasts. In this step, users perform physical inventory operations based on the plan and provide feedback on the results. Inputs are the inventory allocation plan and on-site information displayed on the terminal, and outputs are the results of the actions taken and the feedback thereto. Based on the results, inventory optimization and model accuracy improvements are performed.

[0343] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0344] This invention combines an emotion engine with an inventory management system to optimize inventory allocation while considering the user's emotional state. The system is composed of three elements: a server, a terminal, and a user.

[0345] server

[0346] The server builds a demand forecasting model using machine learning algorithms based on sales data collected from various distribution channels. In addition, the emotion engine processes emotional data obtained from user input and actions and incorporates it into the demand forecasting model. The emotion engine analyzes user text input and voice data using natural language processing and sentiment analysis algorithms.

[0347] terminal

[0348] When the terminal presents the user with an inventory allocation plan, the emotion engine displays information tailored to the user's emotional state. For example, if the user is stressed, the terminal provides concise, real-time information and applies an interface to assist in decision-making.

[0349] User

[0350] Users can view inventory data and demand forecasts through their devices to make decisions for the next steps. Sentimental data obtained from user interactions is fed back to the server, which uses this data to continuously improve its forecasting models and displayed content.

[0351] Specific example

[0352] For example, if the emotion engine detects signs such as a user spending a long time working on the inventory management screen and their input speed slowing down, the server records this as a record, and the next time a similar situation is detected, the terminal will send a simplified notification. Also, if the emotion engine detects a state of excitement when demand for a particular product suddenly increases, the server will immediately recommend high-priority inventory management actions.

[0353] In this way, the system provides a form that supports efficient decision-making for users by performing highly accurate inventory management through the collection of sales data and the analysis of sentiment.

[0354] The following describes the processing flow.

[0355] Step 1:

[0356] The server periodically collects sales data from each distribution channel and stores it in a database. This includes sales history, inventory status, and seasonal sales characteristics. The data is then processed using data cleaning algorithms to make it ready for analysis.

[0357] Step 2:

[0358] The server uses an emotion engine to analyze the user's emotional state from their operation logs and input. The emotion engine uses natural language processing and machine learning models to analyze text and voice input and recognize the user's emotional state.

[0359] Step 3:

[0360] The server combines sales data and user sentiment data to build a demand forecasting model. This model is based on machine learning algorithms and is designed to maximize forecasting accuracy by taking into account the characteristics of each distribution channel.

[0361] Step 4:

[0362] The server develops an inventory allocation plan based on predicted demand. This may involve prioritization based on sentiment data, and optimization algorithms are used to ensure that all allocations are reasonable and meet sales criteria.

[0363] Step 5:

[0364] The terminal visually presents the inventory allocation plan received from the server to the user. It incorporates the results of the emotion engine to provide information that reassures the user. For example, if the user is experiencing high stress levels, the notification will offer an intuitive and simple action.

[0365] Step 6:

[0366] Users use their devices to review the proposed inventory allocation plan and make adjustments as needed. Users also send feedback from their devices to the server, providing information, particularly regarding the effectiveness of sentiment-based responses.

[0367] Step 7:

[0368] The server analyzes user feedback and actual sales results, and based on this, continuously improves its demand forecasting model and sentiment analysis algorithm. As a result, the system's accuracy and efficiency improve over time.

[0369] (Example 2)

[0370] Next, we will describe Example 2. 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".

[0371] Conventional inventory management systems rely solely on historical sales data for demand forecasting, resulting in low accuracy in predicting actual demand fluctuations. Furthermore, they fail to consider the impact of user emotional states on inventory allocation planning decisions, potentially leading to inefficient decision-making. Therefore, a more accurate inventory allocation method that reflects user emotional states is needed.

[0372] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0373] In this invention, the server includes means for collecting sales information from each distribution channel and storing it in an information storage device, means for constructing a structure for making demand forecasts based on the information using machine learning technology, and means for processing user emotional information using an emotional analysis engine and reflecting it in the demand forecasting structure. This enables highly accurate demand forecasting and inventory allocation that takes into account the emotional state of the user.

[0374] "Distribution channels" refer to the routes and networks through which products and services are delivered from producers to consumers.

[0375] "Sales information" refers to information that includes data on product sales and sales status, and is used for forecasting future demand.

[0376] An "information storage device" includes hardware and software for storing collected data and accessing and processing it as needed.

[0377] "Machine learning techniques" is a general term for algorithms and methods that can learn patterns from large amounts of data and perform predictions and classifications.

[0378] "Structure" refers to algorithms and data flows used to process information and build models, and is a crucial element in system design.

[0379] A "sentiment analysis engine" is a software technology that uses natural language processing and speech processing to analyze human emotions and utilize them as data.

[0380] "Users" refer to individuals who operate the system, input data, and make decisions.

[0381] "Emotional information" refers to data that indicates the emotional state of a user, analyzed based on their actions and inputs.

[0382] "Demand forecasting" refers to the process of predicting future demand for a product based on past sales information and current data.

[0383] This invention is a new type of system that combines an inventory management system and an emotion engine, and its core is based on data exchange between a server, a terminal, and a user.

[0384] server

[0385] The server primarily collects sales information from various distribution channels and stores it in a database, which serves as the information storage device. This process involves periodically executing data collection scripts and retrieving sales information via APIs and other data acquisition methods. The collected information is used to build demand forecasting models using machine learning techniques. Specifically, random forest and linear regression models are trained using Python's Scikit-learn. Furthermore, the server utilizes a sentiment analysis engine to analyze text input and audio data from users and extract sentiment information. Natural language processing libraries such as NLTK and Transformers are used for this process. The analyzed sentiment information is then incorporated as new features into the demand forecasting model, resulting in more accurate predictions.

[0386] terminal

[0387] When the terminal presents the inventory allocation plan transmitted from the server to the user, the information display is adjusted based on the results of the sentiment analysis engine. For example, if the user's stress level is detected, the information is provided in a simplified form in real time, and an interface is applied to support necessary decision-making. This makes it easier for the user to understand and make quick decisions.

[0388] User

[0389] Users access inventory data and demand forecasts through their terminals and make decisions for the next steps based on that information. Sentimental information and feedback data obtained through user interaction are fed back to the server and used for continuous improvement of the demand forecasting model and displayed content. In this way, the entire system forms a dynamic improvement cycle that takes user emotions into account, enhancing the user experience.

[0390] Specific example

[0391] For example, if the emotion engine detects a user's level of interest by frequently checking the inventory status of a particular product, it will prioritize displaying inventory data for that product the next time the user checks. In addition, if the analysis reveals that the user is experiencing fatigue due to slow input speed, the device can simplify the information and display only the most important data.

[0392] Examples of prompts for generative AI models

[0393] "Based on this week's sales data and user sentiment trends, please propose an inventory allocation plan for next week."

[0394] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0395] Step 1:

[0396] The server collects sales information from each distribution channel. The input is raw sales data obtained via APIs and other data acquisition methods. The server organizes this data and stores it in an information storage device. Data cleaning is performed to remove inaccurate data before it is stored in the database. The output is a clean set of sales information.

[0397] Step 2:

[0398] The server builds a demand forecasting model using sales information. The input is the set of sales information obtained in the previous step. The server uses Scikit-learn, a machine learning technique, to train the model by applying algorithms such as random forest and linear regression. During this process, distribution and trend analysis is performed, and a model for predicting future demand is output.

[0399] Step 3:

[0400] The server performs user sentiment analysis. The input is text or voice data from the user. The server analyzes this data using a natural language processing library and extracts the user's sentiment information. The sentiment analysis engine identifies emotional states such as positive, negative, or neutral, and provides sentiment information as output.

[0401] Step 4:

[0402] The server incorporates the acquired sentiment information into the demand forecasting model. The inputs are the demand forecasting model and the sentiment information. The server adds the sentiment information as a feature of the model and readjusts it. This readjustment enables highly accurate forecasts that take emotional states into account. The output is the demand forecasting model with sentiment reflected.

[0403] Step 5:

[0404] The terminal develops and presents an inventory allocation plan to the user based on a demand forecasting model that incorporates sentiment from the server. The inputs are the demand forecasting model and inventory data. The terminal analyzes the information and provides the user with an interface for visualization. During this process, the displayed content is adjusted according to the user's emotional state, and the adjusted inventory allocation plan is presented to the user as output.

[0405] Step 6:

[0406] The user reviews the presented inventory allocation plan and makes adjustments as needed. The input is the inventory allocation plan presented by the terminal. User interaction through the interface generates feedback information regarding the plan. The output is the information sent to the server as user feedback data.

[0407] Step 7:

[0408] The server improves the model based on user feedback. The inputs are feedback data and the current demand forecasting model. The server analyzes the feedback and tunes the model parameters as needed. This improves forecasting accuracy, resulting in an improved demand forecasting model as output.

[0409] (Application Example 2)

[0410] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0411] Traditional inventory management systems perform demand forecasting and inventory allocation based on sales data, but they do not take into account the emotional state of users, and therefore cannot be said to have fully achieved optimization of inventory allocation. In particular, when users are experiencing stress or anxiety, the way inventory information is provided may not be appropriate. In such situations, the efficiency of decision-making decreases, and as a result, the accuracy of inventory management is compromised.

[0412] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0413] In this invention, the server includes means for collecting sales data from each distribution channel and storing it in a database, means for constructing a demand forecasting model using a machine learning algorithm based on the data, and means for influencing the forecasted demand based on the user's emotional state. This makes it possible to optimize inventory allocation while taking into account the user's emotional state.

[0414] "Sales data" refers to records of how products were sold through each distribution channel.

[0415] A "database" is a collection of data built to efficiently store, manage, and retrieve collected information.

[0416] A "machine learning algorithm" is a computational method used to build predictive models that improve themselves using collected data.

[0417] "User emotional state" refers to information about the psychological state or changes a user exhibits when operating a system.

[0418] An "inventory allocation plan" is a strategy for efficiently distributing goods to each location based on predicted demand.

[0419] A "display device" is a device used to visually present information to a user.

[0420] "Feedback" is the process of returning actual performance data to the system and using it to improve models and plans.

[0421] The system that implements this application example includes a series of processes to optimize inventory management by taking into account the user's emotional state. A detailed explanation of how the system works is provided below.

[0422] The server collects sales data from each distribution channel and stores it in a database. This prepares the foundational data for demand forecasting. As a machine learning algorithm, "scikit-learn" is used to analyze the sales data and build a demand forecasting model. Furthermore, the sentiment engine extracts sentiment data from user input and actions, and analyzes it using natural language processing libraries such as "NLTK" and "TextBlob". This sentiment data is then reflected in the demand forecasting model.

[0423] The device displays information tailored to the user's emotional state, and if the user is feeling stressed, it presents information in a concise and concise manner. This allows the user to make quick and accurate decisions.

[0424] Based on the inventory data and forecast information displayed on the device, users decide what action to take next. Sentiment data obtained from user interactions is constantly fed back to the server and used to improve the accuracy of the model and optimize the displayed content.

[0425] A concrete example would be a scenario where an employee working at a logistics center gives a voice command via their smartphone, saying, "Tell me the latest demand forecast." In this case, if the system senses anxiety from the voice, it will concisely display only the most important information visually. An example of a prompt message would be, "Infer anxiety from the voice and display inventory information accordingly."

[0426] This system streamlines the inventory management process at logistics centers and reduces the psychological burden on users.

[0427] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0428] Step 1:

[0429] The server collects sales data from each distribution channel and stores it in a database. This sales data contains historical information about when and through which channels products were sold. Sales logs from each store and distributor are provided as input, and this data is organized and stored in the database as output. Specifically, the server performs data insertion into the database and consistency checks.

[0430] Step 2:

[0431] The server uses collected sales data to build a demand forecasting model based on a machine learning algorithm. Sales performance and distribution channel characteristics are provided as input data, and the demand forecasting model is generated as output. This process utilizes "scikit-learn" for data preprocessing, algorithm selection, model training, and evaluation of prediction accuracy.

[0432] Step 3:

[0433] The server analyzes voice or text data obtained from user input and actions using natural language processing tools such as "NLTK" and "TextBlob" to evaluate the user's emotional state. Voice data or text chat is provided as input, and the user's emotional state is obtained as output. Specific operations include text conversion, sentiment analysis, and identification of emotional tendencies.

[0434] Step 4:

[0435] The server updates the demand forecasting model based on the sentiment data obtained from the sentiment engine. Sentiment data is input, and a new demand forecast reflecting the sentiment is created as output. In this process, the model is retrained to incorporate the sentiment data.

[0436] Step 5:

[0437] The terminal receives the inventory allocation plan provided by the server and displays it to the user. Here, the information is simplified according to the user's emotional state. Inventory allocation information is received as input, and a visual display tailored to the user's emotions is produced as output. Specifically, filtering and highlighting of the displayed content are performed.

[0438] Step 6:

[0439] The user makes decisions based on the presented inventory allocation plan and feeds the results back to the server. The inventory allocation plan is displayed as input, and feedback data is generated as output. Specifically, the actions decided by the user are sent to the server and used for future model improvements of the system.

[0440] The specific processing unit 290 transmits the result of the specific processing 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 unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0441] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. 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. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0442] 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.

[0443] [Third Embodiment]

[0444] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0445] 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.

[0446] 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).

[0447] 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, and display 343 are also connected to the bus 52.

[0448] 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.

[0449] 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).

[0450] 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.

[0451] 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.

[0452] 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.

[0453] 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.

[0454] 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.

[0455] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. 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".

[0456] This invention provides a system for achieving efficient and highly accurate inventory management, and a specific embodiment thereof is shown. The system mainly consists of three elements: a server, a terminal, and a user.

[0457] server

[0458] The server is responsible for collecting sales data from each distribution channel and storing it in a database. This allows for demand forecasting tailored to the characteristics of each sales channel, based on the data accumulated for each channel. The server uses machine learning algorithms to build models and perform highly accurate demand forecasts. Furthermore, the server develops an inventory allocation plan based on the predicted demand and provides that plan to the terminals.

[0459] terminal

[0460] The terminal visually presents the inventory allocation plan provided by the server to the user. Using visualization tools such as charts and graphs, the terminal helps the user intuitively understand inventory flow and demand forecasts. This allows the terminal to support the user in making quick decisions.

[0461] User

[0462] Users receive inventory information via their terminals and incorporate it into their actual on-site operations. Users also provide sales performance as feedback to the system, and the server uses this feedback to continuously improve the demand forecasting model. As a result, the system learns over time, enabling more accurate and faster inventory management.

[0463] Specific example

[0464] For example, if a sudden surge in demand for used devices is predicted at a particular store, the server immediately processes this information and instructs the user via the device on the optimal inventory quantity. Based on this instruction, the user secures the necessary inventory and provides it to the store. After the sale is completed, the user provides feedback on the performance data, and the server uses this data to improve the accuracy of predictions for the next cycle.

[0465] In this way, the system automates everything from sales data collection to feedback, reducing inventory surpluses and shortages, and enabling efficient inventory management tailored to the characteristics of each distribution channel.

[0466] The following describes the processing flow.

[0467] Step 1:

[0468] The server periodically collects sales data from each distribution channel. The collected data is stored in a database via queries, and missing or outlier values ​​are corrected through a data cleaning process.

[0469] Step 2:

[0470] The server applies machine learning algorithms to clean sales data to build a demand forecasting model. The model is optimized to take into account channel-specific patterns and seasonality, and is validated on a test dataset to evaluate its accuracy.

[0471] Step 3:

[0472] The server uses a trained demand forecasting model to predict demand for the next cycle. New data inputs are also incorporated into the demand forecast, and the forecast results are stored in a database.

[0473] Step 4:

[0474] The server develops an inventory allocation plan based on predicted demand. Here, optimization algorithms such as linear programming are used to calculate the optimal inventory allocation, taking into account constraints such as warehouse capacity limits and transportation costs.

[0475] Step 5:

[0476] The server sends inventory allocation plans and forecast data to the terminal. The terminal provides an interface to visually present information to the user, allowing them to intuitively understand inventory flow and demand forecasts.

[0477] Step 6:

[0478] Users check inventory information provided from their terminals and perform actual inventory management operations. They issue necessary inventory transfer and procurement instructions, and prepare for sales.

[0479] Step 7:

[0480] Users provide feedback on their actual sales performance to the server via their devices. The server uses this feedback data to readjust its demand forecasting model, enabling more accurate predictions in the next cycle.

[0481] (Example 1)

[0482] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0483] Inventory management systems are required to respond immediately to fluctuations in sales information, make more accurate demand forecasts, and achieve efficient inventory allocation. However, existing systems fail to adequately consider the characteristics of each distribution channel, which can lead to decreased forecast accuracy and inventory shortages or surpluses. Furthermore, there is a lack of established methods for effectively utilizing feedback information to improve models, making rapid decision-making difficult.

[0484] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0485] In this invention, the server includes means for collecting sales information along the distribution route and storing it in an information storage device, means for constructing a forecasting model for demand forecasting based on the information using machine learning techniques, and means for establishing an inventory allocation plan based on the forecasted demand. This makes it possible to improve forecasting accuracy by considering the characteristics of each distribution route and using a generative AI model, and to achieve optimized inventory allocation that takes into account the capacity limitations of storage facilities and logistics costs.

[0486] "Distribution channels" refer to the series of paths or channels involved in the process by which goods or services are delivered from producers to consumers.

[0487] "Sales information" refers to data related to the sales of goods and services, including purchase date and time, sales quantity, and customer information.

[0488] An "information storage device" refers to a storage medium or storage system that can store data and retrieve it as needed.

[0489] "Demand forecasting" refers to the analysis and methods used to predict the size and fluctuations of future demand.

[0490] A "predictive model" refers to a mathematical or algorithmic structure built to predict future events or trends based on data.

[0491] "Machine learning techniques" refer to algorithms and processes that learn patterns from data and use them to make future predictions and classifications.

[0492] An "inventory allocation plan" refers to a plan or strategy for optimally positioning and allocating product inventory to each distribution point based on demand forecasts.

[0493] A "generative AI model" refers to a model that uses artificial intelligence technology to learn from data and is generated to respond to new data or specific conditions.

[0494] A "storage facility" refers to a physical structure or location designed for the storage of goods or materials.

[0495] An "optimization method" refers to a mathematical or algorithmic technique for utilizing resources and conditions in the most efficient way to achieve a specific objective.

[0496] Modes for carrying out the invention

[0497] This invention provides an efficient and highly accurate inventory management system. The specific implementation of this system is described below.

[0498] server

[0499] The server's role is to collect sales information from each distribution channel and store it in a data storage device. Specifically, it retrieves data from multiple sales channels via APIs. This data includes product identification information, the number of units sold, and the sales date. The data is stored using a database management system (e.g., MySQL). The server preprocesses and cleanses the data using Python or R, and then builds a demand forecasting model using machine learning techniques (e.g., TensorFlow or Scikit-learn). Furthermore, the server develops an optimal inventory allocation plan based on this forecast data and provides it to the terminal.

[0500] terminal

[0501] The terminal functions to visually present inventory allocation plans provided by the server to the user. Visualization software such as Microsoft Power BI and Tableau is used to ensure intuitive understanding for the user. Inventory flows and projected demand fluctuations can be displayed in graph and dashboard formats. This allows the terminal to help users make quick and effective decisions.

[0502] User

[0503] Users adjust and replenish inventory based on information provided through their terminals. Furthermore, users input sales performance data into the system as feedback. This includes sales figures, return information, and customer feedback. This feedback data is sent to a server and used to continuously improve the demand forecasting model.

[0504] Specific example

[0505] For example, if a sudden surge in demand for a new product is predicted, the server immediately processes the information and plans the optimal inventory allocation for each store. This information is notified to the user via a terminal, allowing the user to quickly secure the necessary inventory based on the instructions. After sales are complete, the user provides feedback on the performance data, which the server uses to improve future forecasts.

[0506] Example of a prompt

[0507] "Enter new sales data into the server and forecast inventory demand for the next month. Visualize the results, send them to the terminal, and improve the model based on user feedback."

[0508] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0509] Step 1:

[0510] The server collects sales information from each distribution channel and stores it in its data storage device. The input at this stage is raw data from the sales channels, specifically sales figures, dates, and product identification information obtained via APIs. This data is structured and stored using a database management system. The output is organized and stored in a format suitable for subsequent data analysis. Specifically, the server periodically calls APIs to retrieve and store new data.

[0511] Step 2:

[0512] The server preprocesses the stored data. The input is the sales information organized in Step 1. Data processing includes imputing missing values ​​and handling outliers, and is performed using Python's data analysis library. The output is a clean and consistent dataset, which is used to build the predictive model. Specifically, the server generates a data frame and performs the necessary data cleansing.

[0513] Step 3:

[0514] The server builds a demand forecasting model using pre-processed data. The input is a clean dataset, and the model is trained using a machine learning framework. By applying a generative AI model, future demand forecasts are made based on past sales trends. The output is the demand forecast result, which serves as the basis for inventory allocation planning. Specifically, the server optimizes the model's hyperparameters and executes the forecasting algorithm.

[0515] Step 4:

[0516] The server develops an inventory allocation plan based on predicted demand. The input is the demand forecast, and an optimization algorithm is used to calculate the optimal inventory allocation to each sales channel. Factors to consider include distribution costs and warehouse capacity. The output is a specific inventory allocation plan for each sales channel. In practice, the server determines the optimal inventory allocation using linear programming and other optimization techniques.

[0517] Step 5:

[0518] The terminal visually displays the inventory allocation plan received from the server. The input is the inventory allocation plan provided by the server, and the terminal provides information to the user using a visualization tool. The output is intuitive visual information presented in a dashboard format. Specifically, the terminal generates graphs and charts and organizes and displays the data in a way that is easy for the user to understand.

[0519] Step 6:

[0520] The user adjusts inventory based on terminal information and inputs sales performance data as feedback. Input data includes sales volume, return information, and other performance data. The output is improved feedback data, contributing to improved future forecast accuracy. Specifically, the user supports the system's learning process by operating the terminal and accurately inputting feedback information.

[0521] (Application Example 1)

[0522] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0523] In logistics operations, improving the efficiency of inventory management and the accuracy of demand forecasting are crucial challenges. Traditional systems often fail to collect sales data quickly and accurately, leading to inventory shortages and surpluses, as well as increased costs. Furthermore, it is difficult for on-site users to easily understand inventory information and respond promptly.

[0524] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0525] In this invention, the server includes means for collecting sales information and storing it in an information storage device, means for constructing a demand forecasting model using a machine learning algorithm, and means for designing an inventory allocation plan. This enables rapid collection and analysis of sales information, and efficient inventory management through highly accurate demand forecasting. Furthermore, by visually representing inventory liquidity through an information terminal and using voice support technology, users can intuitively understand the information and make quick decisions.

[0526] "Route" is a term that refers to the points or paths that goods or information take to circulate.

[0527] An "information storage device" is an electronic device or system for accumulating various types of collected data and storing them in a format that makes them easily accessible when needed.

[0528] A "machine learning algorithm" is a method or computational procedure that allows a computer to learn patterns from data and perform predictions and classifications.

[0529] An "inventory allocation plan" is a strategic plan for appropriately allocating inventory of goods and materials to each location based on demand forecasts.

[0530] A "display device" is an electronic device that visually displays information and assists users in confirming and managing it.

[0531] "Users" refer to individuals who operate systems and devices, and who utilize information to perform tasks and make decisions.

[0532] "Voice assistance technology" is a technology that uses voice to give instructions and obtain information, enabling users to access information hands-free.

[0533] A "generative AI model" is an artificial intelligence model that learns from given data and prompts and is generated to address new problems.

[0534] A "prompt statement" refers to a document or command used when inputting instructions or data into an AI model.

[0535] The system that implements this application consists of three elements: a server, a terminal, and a user. Its aim is to create an efficient environment for inventory management.

[0536] The server collects sales information and stores it in an information storage device. The collected data is used to build a demand forecasting model using machine learning algorithms. Specifically, Flask is used for data communication, and TensorFlow is used to train the forecasting model. Based on the built model, the server designs an inventory allocation plan, which is then sent to terminals via information terminals.

[0537] The terminal is responsible for visually presenting the inventory allocation plan received from the server to the user. The terminal has an application built using React Native installed, enabling data visualization using Charts.js. Furthermore, it utilizes Google Cloud Speech-to-Text as a voice assistance technology, providing users with the ability to access information hands-free.

[0538] Users use their devices to check and manage inventory information and allocation plans. The decisions made by users are sent to the server as feedback. This feedback is used to improve the model in the next cycle, enhancing the accuracy of the generated AI model.

[0539] To give a specific example, if a sudden fluctuation in demand for a particular product is predicted at a logistics center, the server immediately processes that information and instructs the terminal to allocate inventory optimally. Based on these instructions, the user optimizes inventory, enabling efficient logistics without surpluses or shortages.

[0540] An example of a prompt message for a generative AI model might be: "Create a program that forecasts demand for a specific product category and displays the results in a smartphone app."

[0541] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0542] Step 1:

[0543] The server collects sales information from each distribution channel and stores it in the information storage device. In this step, the server collects sales information into a database and organizes and stores the data. The input is raw sales data from each distribution channel, and the output is organized sales data stored in the information storage device. Data normalization and duplicate removal are performed.

[0544] Step 2:

[0545] The server uses accumulated sales data to execute a machine learning algorithm and build a demand forecasting model. In this step, the model is trained using tools such as TensorFlow. The input is organized sales data stored in an information storage device, and the output is a trained model specifically designed for demand forecasting. After data preprocessing, the model performs pattern recognition and generates forecast data.

[0546] Step 3:

[0547] The server predicts the demand for each product based on the built model and develops an inventory allocation plan. Here, the server performs predictions using the trained model and calculates the optimal inventory allocation based on the results. The input is the trained demand forecasting model and newly collected sales data, and the output is the inventory allocation plan. The forecasting algorithm identifies the supply-demand gap and automatically generates the allocation plan.

[0548] Step 4:

[0549] The terminal visually presents the inventory allocation plan received from the server to the user. In this step, the terminal uses Charts.js visualization via a React Native application. The input is the inventory allocation plan sent from the server, and the output is information presented visually and audibly on the screen and through voice assistance technologies. The terminal uses a voice assistant to inform the user of important information.

[0550] Step 5:

[0551] Users review inventory allocation plans via a terminal and take actions based on on-site conditions. They also send feedback to the server to improve the accuracy of future forecasts. In this step, users perform physical inventory operations based on the plan and provide feedback on the results. Inputs are the inventory allocation plan and on-site information displayed on the terminal, and outputs are the results of the actions taken and the feedback thereto. Based on the results, inventory optimization and model accuracy improvements are performed.

[0552] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0553] This invention combines an emotion engine with an inventory management system to optimize inventory allocation while considering the user's emotional state. The system is composed of three elements: a server, a terminal, and a user.

[0554] server

[0555] The server builds a demand forecasting model using machine learning algorithms based on sales data collected from various distribution channels. In addition, the emotion engine processes emotional data obtained from user input and actions and incorporates it into the demand forecasting model. The emotion engine analyzes user text input and voice data using natural language processing and sentiment analysis algorithms.

[0556] terminal

[0557] When the terminal presents the user with an inventory allocation plan, the emotion engine displays information tailored to the user's emotional state. For example, if the user is stressed, the terminal provides concise, real-time information and applies an interface to assist in decision-making.

[0558] User

[0559] Users can view inventory data and demand forecasts through their devices to make decisions for the next steps. Sentimental data obtained from user interactions is fed back to the server, which uses this data to continuously improve its forecasting models and displayed content.

[0560] Specific example

[0561] For example, if the emotion engine detects signs such as a user spending a long time working on the inventory management screen and their input speed slowing down, the server records this as a record, and the next time a similar situation is detected, the terminal will send a simplified notification. Also, if the emotion engine detects a state of excitement when demand for a particular product suddenly increases, the server will immediately recommend high-priority inventory management actions.

[0562] In this way, the system provides a form that supports efficient decision-making for users by performing highly accurate inventory management through the collection of sales data and the analysis of sentiment.

[0563] The following describes the processing flow.

[0564] Step 1:

[0565] The server periodically collects sales data from each distribution channel and stores it in a database. This includes sales history, inventory status, and seasonal sales characteristics. The data is then processed using data cleaning algorithms to make it ready for analysis.

[0566] Step 2:

[0567] The server uses an emotion engine to analyze the user's emotional state from their operation logs and input. The emotion engine uses natural language processing and machine learning models to analyze text and voice input and recognize the user's emotional state.

[0568] Step 3:

[0569] The server combines sales data and user sentiment data to build a demand forecasting model. This model is based on machine learning algorithms and is designed to maximize forecasting accuracy by taking into account the characteristics of each distribution channel.

[0570] Step 4:

[0571] The server develops an inventory allocation plan based on predicted demand. This may involve prioritization based on sentiment data, and optimization algorithms are used to ensure that all allocations are reasonable and meet sales criteria.

[0572] Step 5:

[0573] The terminal visually presents the inventory allocation plan received from the server to the user. It incorporates the results of the emotion engine to provide information that reassures the user. For example, if the user is experiencing high stress levels, the notification will offer an intuitive and simple action.

[0574] Step 6:

[0575] Users use their devices to review the proposed inventory allocation plan and make adjustments as needed. Users also send feedback from their devices to the server, providing information, particularly regarding the effectiveness of sentiment-based responses.

[0576] Step 7:

[0577] The server analyzes user feedback and actual sales results, and based on this, continuously improves its demand forecasting model and sentiment analysis algorithm. As a result, the system's accuracy and efficiency improve over time.

[0578] (Example 2)

[0579] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0580] Conventional inventory management systems rely solely on historical sales data for demand forecasting, resulting in low accuracy in predicting actual demand fluctuations. Furthermore, they fail to consider the impact of user emotional states on inventory allocation planning decisions, potentially leading to inefficient decision-making. Therefore, a more accurate inventory allocation method that reflects user emotional states is needed.

[0581] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0582] In this invention, the server includes means for collecting sales information from each distribution channel and storing it in an information storage device, means for constructing a structure for making demand forecasts based on the information using machine learning technology, and means for processing user emotional information using an emotional analysis engine and reflecting it in the demand forecasting structure. This enables highly accurate demand forecasting and inventory allocation that takes into account the emotional state of the user.

[0583] "Distribution channels" refer to the routes and networks through which products and services are delivered from producers to consumers.

[0584] "Sales information" refers to information that includes data on product sales and sales status, and is used for forecasting future demand.

[0585] An "information storage device" includes hardware and software for storing collected data and accessing and processing it as needed.

[0586] "Machine learning techniques" is a general term for algorithms and methods that can learn patterns from large amounts of data and perform predictions and classifications.

[0587] "Structure" refers to algorithms and data flows used to process information and build models, and is a crucial element in system design.

[0588] A "sentiment analysis engine" is a software technology that uses natural language processing and speech processing to analyze human emotions and utilize them as data.

[0589] "Users" refer to individuals who operate the system, input data, and make decisions.

[0590] "Emotional information" refers to data that indicates the emotional state of a user, analyzed based on their actions and inputs.

[0591] "Demand forecasting" refers to the process of predicting future demand for a product based on past sales information and current data.

[0592] This invention is a new type of system that combines an inventory management system and an emotion engine, and its core is based on data exchange between a server, a terminal, and a user.

[0593] server

[0594] The server primarily collects sales information from various distribution channels and stores it in a database, which serves as the information storage device. This process involves periodically executing data collection scripts and retrieving sales information via APIs and other data acquisition methods. The collected information is used to build demand forecasting models using machine learning techniques. Specifically, random forest and linear regression models are trained using Python's Scikit-learn. Furthermore, the server utilizes a sentiment analysis engine to analyze text input and audio data from users and extract sentiment information. Natural language processing libraries such as NLTK and Transformers are used for this process. The analyzed sentiment information is then incorporated as new features into the demand forecasting model, resulting in more accurate predictions.

[0595] terminal

[0596] When the terminal presents the inventory allocation plan transmitted from the server to the user, the information display is adjusted based on the results of the sentiment analysis engine. For example, if the user's stress level is detected, the information is provided in a simplified form in real time, and an interface is applied to support necessary decision-making. This makes it easier for the user to understand and make quick decisions.

[0597] User

[0598] Users access inventory data and demand forecasts through their terminals and make decisions for the next steps based on that information. Sentimental information and feedback data obtained through user interaction are fed back to the server and used for continuous improvement of the demand forecasting model and displayed content. In this way, the entire system forms a dynamic improvement cycle that takes user emotions into account, enhancing the user experience.

[0599] Specific example

[0600] For example, if the emotion engine detects a user's level of interest by frequently checking the inventory status of a particular product, it will prioritize displaying inventory data for that product the next time the user checks. In addition, if the analysis reveals that the user is experiencing fatigue due to slow input speed, the device can simplify the information and display only the most important data.

[0601] Examples of prompts for generative AI models

[0602] "Based on this week's sales data and user sentiment trends, please propose an inventory allocation plan for next week."

[0603] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0604] Step 1:

[0605] The server collects sales information from each distribution channel. The input is raw sales data obtained via APIs and other data acquisition methods. The server organizes this data and stores it in an information storage device. Data cleaning is performed to remove inaccurate data before it is stored in the database. The output is a clean set of sales information.

[0606] Step 2:

[0607] The server builds a demand forecasting model using sales information. The input is the set of sales information obtained in the previous step. The server uses Scikit-learn, a machine learning technique, to train the model by applying algorithms such as random forest and linear regression. During this process, distribution and trend analysis is performed, and a model for predicting future demand is output.

[0608] Step 3:

[0609] The server performs user sentiment analysis. The input is text or voice data from the user. The server analyzes this data using a natural language processing library and extracts the user's sentiment information. The sentiment analysis engine identifies emotional states such as positive, negative, or neutral, and provides sentiment information as output.

[0610] Step 4:

[0611] The server incorporates the acquired sentiment information into the demand forecasting model. The inputs are the demand forecasting model and the sentiment information. The server adds the sentiment information as a feature of the model and readjusts it. This readjustment enables highly accurate forecasts that take emotional states into account. The output is the demand forecasting model with sentiment reflected.

[0612] Step 5:

[0613] The terminal develops and presents an inventory allocation plan to the user based on a demand forecasting model that incorporates sentiment from the server. The inputs are the demand forecasting model and inventory data. The terminal analyzes the information and provides the user with an interface for visualization. During this process, the displayed content is adjusted according to the user's emotional state, and the adjusted inventory allocation plan is presented to the user as output.

[0614] Step 6:

[0615] The user reviews the presented inventory allocation plan and makes adjustments as needed. The input is the inventory allocation plan presented by the terminal. User interaction through the interface generates feedback information regarding the plan. The output is the information sent to the server as user feedback data.

[0616] Step 7:

[0617] The server improves the model based on user feedback. The inputs are feedback data and the current demand forecasting model. The server analyzes the feedback and tunes the model parameters as needed. This improves forecasting accuracy, resulting in an improved demand forecasting model as output.

[0618] (Application Example 2)

[0619] Next, we will explain Application Example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0620] Traditional inventory management systems perform demand forecasting and inventory allocation based on sales data, but they do not take into account the emotional state of users, and therefore cannot be said to have fully achieved optimization of inventory allocation. In particular, when users are experiencing stress or anxiety, the way inventory information is provided may not be appropriate. In such situations, the efficiency of decision-making decreases, and as a result, the accuracy of inventory management is compromised.

[0621] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0622] In this invention, the server includes means for collecting sales data from each distribution channel and storing it in a database, means for constructing a demand forecasting model using a machine learning algorithm based on the data, and means for influencing the forecasted demand based on the user's emotional state. This makes it possible to optimize inventory allocation while taking into account the user's emotional state.

[0623] "Sales data" refers to records of how products were sold through each distribution channel.

[0624] A "database" is a collection of data built to efficiently store, manage, and retrieve collected information.

[0625] A "machine learning algorithm" is a computational method used to build predictive models that improve themselves using collected data.

[0626] "User emotional state" refers to information about the psychological state or changes a user exhibits when operating a system.

[0627] An "inventory allocation plan" is a strategy for efficiently distributing goods to each location based on predicted demand.

[0628] A "display device" is a device used to visually present information to a user.

[0629] "Feedback" is the process of returning actual performance data to the system and using it to improve models and plans.

[0630] The system that implements this application example includes a series of processes to optimize inventory management by taking into account the user's emotional state. A detailed explanation of how the system works is provided below.

[0631] The server collects sales data from each distribution channel and stores it in a database. This prepares the foundational data for demand forecasting. As a machine learning algorithm, "scikit-learn" is used to analyze the sales data and build a demand forecasting model. Furthermore, the sentiment engine extracts sentiment data from user input and actions, and analyzes it using natural language processing libraries such as "NLTK" and "TextBlob". This sentiment data is then reflected in the demand forecasting model.

[0632] The device displays information tailored to the user's emotional state, and if the user is feeling stressed, it presents information in a concise and concise manner. This allows the user to make quick and accurate decisions.

[0633] Based on the inventory data and forecast information displayed on the device, users decide what action to take next. Sentiment data obtained from user interactions is constantly fed back to the server and used to improve the accuracy of the model and optimize the displayed content.

[0634] A concrete example would be a scenario where an employee working at a logistics center gives a voice command via their smartphone, saying, "Tell me the latest demand forecast." In this case, if the system senses anxiety from the voice, it will concisely display only the most important information visually. An example of a prompt message would be, "Infer anxiety from the voice and display inventory information accordingly."

[0635] This system streamlines the inventory management process at logistics centers and reduces the psychological burden on users.

[0636] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0637] Step 1:

[0638] The server collects sales data from each distribution channel and stores it in a database. This sales data contains historical information about when and through which channels products were sold. Sales logs from each store and distributor are provided as input, and this data is organized and stored in the database as output. Specifically, the server performs data insertion into the database and consistency checks.

[0639] Step 2:

[0640] The server uses collected sales data to build a demand forecasting model based on a machine learning algorithm. Sales performance and distribution channel characteristics are provided as input data, and the demand forecasting model is generated as output. This process utilizes "scikit-learn" for data preprocessing, algorithm selection, model training, and evaluation of prediction accuracy.

[0641] Step 3:

[0642] The server analyzes voice or text data obtained from user input and actions using natural language processing tools such as "NLTK" and "TextBlob" to evaluate the user's emotional state. Voice data or text chat is provided as input, and the user's emotional state is obtained as output. Specific operations include text conversion, sentiment analysis, and identification of emotional tendencies.

[0643] Step 4:

[0644] The server updates the demand forecasting model based on the sentiment data obtained from the sentiment engine. Sentiment data is input, and a new demand forecast reflecting the sentiment is created as output. In this process, the model is retrained to incorporate the sentiment data.

[0645] Step 5:

[0646] The terminal receives the inventory allocation plan provided by the server and displays it to the user. Here, the information is simplified according to the user's emotional state. Inventory allocation information is received as input, and a visual display tailored to the user's emotions is produced as output. Specifically, filtering and highlighting of the displayed content are performed.

[0647] Step 6:

[0648] The user makes decisions based on the presented inventory allocation plan and feeds the results back to the server. The inventory allocation plan is displayed as input, and feedback data is generated as output. Specifically, the actions decided by the user are sent to the server and used for future model improvements of the system.

[0649] The specific processing unit 290 transmits the result of the specific processing 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 unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0650] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. 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. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0651] 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.

[0652] [Fourth Embodiment]

[0653] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0654] 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.

[0655] 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).

[0656] 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, and controlled object 443 are also connected to the bus 52.

[0657] 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.

[0658] 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).

[0659] 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.

[0660] 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.

[0661] 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.

[0662] 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.

[0663] 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.

[0664] 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.

[0665] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0666] This invention provides a system for achieving efficient and highly accurate inventory management, and a specific embodiment thereof is shown. The system mainly consists of three elements: a server, a terminal, and a user.

[0667] server

[0668] The server is responsible for collecting sales data from each distribution channel and storing it in a database. This allows for demand forecasting tailored to the characteristics of each sales channel, based on the data accumulated for each channel. The server uses machine learning algorithms to build models and perform highly accurate demand forecasts. Furthermore, the server develops an inventory allocation plan based on the predicted demand and provides that plan to the terminals.

[0669] terminal

[0670] The terminal visually presents the inventory allocation plan provided by the server to the user. Using visualization tools such as charts and graphs, the terminal helps the user intuitively understand inventory flow and demand forecasts. This allows the terminal to support the user in making quick decisions.

[0671] User

[0672] Users receive inventory information via their terminals and incorporate it into their actual on-site operations. Users also provide sales performance as feedback to the system, and the server uses this feedback to continuously improve the demand forecasting model. As a result, the system learns over time, enabling more accurate and faster inventory management.

[0673] Specific example

[0674] For example, if a sudden surge in demand for used devices is predicted at a particular store, the server immediately processes this information and instructs the user via the device on the optimal inventory quantity. Based on this instruction, the user secures the necessary inventory and provides it to the store. After the sale is completed, the user provides feedback on the performance data, and the server uses this data to improve the accuracy of predictions for the next cycle.

[0675] In this way, the system automates everything from sales data collection to feedback, reducing inventory surpluses and shortages, and enabling efficient inventory management tailored to the characteristics of each distribution channel.

[0676] The following describes the processing flow.

[0677] Step 1:

[0678] The server periodically collects sales data from each distribution channel. The collected data is stored in a database via queries, and missing or outlier values ​​are corrected through a data cleaning process.

[0679] Step 2:

[0680] The server applies machine learning algorithms to clean sales data to build a demand forecasting model. The model is optimized to take into account channel-specific patterns and seasonality, and is validated on a test dataset to evaluate its accuracy.

[0681] Step 3:

[0682] The server uses a trained demand forecasting model to predict demand for the next cycle. New data inputs are also incorporated into the demand forecast, and the forecast results are stored in a database.

[0683] Step 4:

[0684] The server develops an inventory allocation plan based on predicted demand. Here, optimization algorithms such as linear programming are used to calculate the optimal inventory allocation, taking into account constraints such as warehouse capacity limits and transportation costs.

[0685] Step 5:

[0686] The server sends inventory allocation plans and forecast data to the terminal. The terminal provides an interface to visually present information to the user, allowing them to intuitively understand inventory flow and demand forecasts.

[0687] Step 6:

[0688] Users check inventory information provided from their terminals and perform actual inventory management operations. They issue necessary inventory transfer and procurement instructions, and prepare for sales.

[0689] Step 7:

[0690] Users provide feedback on their actual sales performance to the server via their devices. The server uses this feedback data to readjust its demand forecasting model, enabling more accurate predictions in the next cycle.

[0691] (Example 1)

[0692] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0693] Inventory management systems are required to respond immediately to fluctuations in sales information, make more accurate demand forecasts, and achieve efficient inventory allocation. However, existing systems fail to adequately consider the characteristics of each distribution channel, which can lead to decreased forecast accuracy and inventory shortages or surpluses. Furthermore, there is a lack of established methods for effectively utilizing feedback information to improve models, making rapid decision-making difficult.

[0694] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0695] In this invention, the server includes means for collecting sales information along the distribution route and storing it in an information storage device, means for constructing a forecasting model for demand forecasting based on the information using machine learning techniques, and means for establishing an inventory allocation plan based on the forecasted demand. This makes it possible to improve forecasting accuracy by considering the characteristics of each distribution route and using a generative AI model, and to achieve optimized inventory allocation that takes into account the capacity limitations of storage facilities and logistics costs.

[0696] "Distribution channels" refer to the series of paths or channels involved in the process by which goods or services are delivered from producers to consumers.

[0697] "Sales information" refers to data related to the sales of goods and services, including purchase date and time, sales quantity, and customer information.

[0698] An "information storage device" refers to a storage medium or storage system that can store data and retrieve it as needed.

[0699] "Demand forecasting" refers to the analysis and methods used to predict the size and fluctuations of future demand.

[0700] A "predictive model" refers to a mathematical or algorithmic structure built to predict future events or trends based on data.

[0701] "Machine learning techniques" refer to algorithms and processes that learn patterns from data and use them to make future predictions and classifications.

[0702] An "inventory allocation plan" refers to a plan or strategy for optimally positioning and allocating product inventory to each distribution point based on demand forecasts.

[0703] A "generative AI model" refers to a model that uses artificial intelligence technology to learn from data and is generated to respond to new data or specific conditions.

[0704] A "storage facility" refers to a physical structure or location designed for the storage of goods or materials.

[0705] An "optimization method" refers to a mathematical or algorithmic technique for utilizing resources and conditions in the most efficient way to achieve a specific objective.

[0706] Modes for carrying out the invention

[0707] This invention provides an efficient and highly accurate inventory management system. The specific implementation of this system is described below.

[0708] server

[0709] The server's role is to collect sales information from each distribution channel and store it in a data storage device. Specifically, it retrieves data from multiple sales channels via APIs. This data includes product identification information, the number of units sold, and the sales date. The data is stored using a database management system (e.g., MySQL). The server preprocesses and cleanses the data using Python or R, and then builds a demand forecasting model using machine learning techniques (e.g., TensorFlow or Scikit-learn). Furthermore, the server develops an optimal inventory allocation plan based on this forecast data and provides it to the terminal.

[0710] terminal

[0711] The terminal functions to visually present inventory allocation plans provided by the server to the user. Visualization software such as Microsoft Power BI and Tableau is used to ensure intuitive understanding for the user. Inventory flows and projected demand fluctuations can be displayed in graph and dashboard formats. This allows the terminal to help users make quick and effective decisions.

[0712] User

[0713] Users adjust and replenish inventory based on information provided through their terminals. Furthermore, users input sales performance data into the system as feedback. This includes sales figures, return information, and customer feedback. This feedback data is sent to a server and used to continuously improve the demand forecasting model.

[0714] Specific example

[0715] For example, if a sudden surge in demand for a new product is predicted, the server immediately processes the information and plans the optimal inventory allocation for each store. This information is notified to the user via a terminal, allowing the user to quickly secure the necessary inventory based on the instructions. After sales are complete, the user provides feedback on the performance data, which the server uses to improve future forecasts.

[0716] Example of a prompt

[0717] "Enter new sales data into the server and forecast inventory demand for the next month. Visualize the results, send them to the terminal, and improve the model based on user feedback."

[0718] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0719] Step 1:

[0720] The server collects sales information from each distribution channel and stores it in its data storage device. The input at this stage is raw data from the sales channels, specifically sales figures, dates, and product identification information obtained via APIs. This data is structured and stored using a database management system. The output is organized and stored in a format suitable for subsequent data analysis. Specifically, the server periodically calls APIs to retrieve and store new data.

[0721] Step 2:

[0722] The server preprocesses the stored data. The input is the sales information organized in Step 1. Data processing includes imputing missing values ​​and handling outliers, and is performed using Python's data analysis library. The output is a clean and consistent dataset, which is used to build the predictive model. Specifically, the server generates a data frame and performs the necessary data cleansing.

[0723] Step 3:

[0724] The server builds a demand forecasting model using pre-processed data. The input is a clean dataset, and the model is trained using a machine learning framework. By applying a generative AI model, future demand forecasts are made based on past sales trends. The output is the demand forecast result, which serves as the basis for inventory allocation planning. Specifically, the server optimizes the model's hyperparameters and executes the forecasting algorithm.

[0725] Step 4:

[0726] The server develops an inventory allocation plan based on predicted demand. The input is the demand forecast, and an optimization algorithm is used to calculate the optimal inventory allocation to each sales channel. Factors to consider include distribution costs and warehouse capacity. The output is a specific inventory allocation plan for each sales channel. In practice, the server determines the optimal inventory allocation using linear programming and other optimization techniques.

[0727] Step 5:

[0728] The terminal visually displays the inventory allocation plan received from the server. The input is the inventory allocation plan provided by the server, and the terminal provides information to the user using a visualization tool. The output is intuitive visual information presented in a dashboard format. Specifically, the terminal generates graphs and charts and organizes and displays the data in a way that is easy for the user to understand.

[0729] Step 6:

[0730] The user adjusts inventory based on terminal information and inputs sales performance data as feedback. Input data includes sales volume, return information, and other performance data. The output is improved feedback data, contributing to improved future forecast accuracy. Specifically, the user supports the system's learning process by operating the terminal and accurately inputting feedback information.

[0731] (Application Example 1)

[0732] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0733] In logistics operations, improving the efficiency of inventory management and the accuracy of demand forecasting are crucial challenges. Traditional systems often fail to collect sales data quickly and accurately, leading to inventory shortages and surpluses, as well as increased costs. Furthermore, it is difficult for on-site users to easily understand inventory information and respond promptly.

[0734] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0735] In this invention, the server includes means for collecting sales information and storing it in an information storage device, means for constructing a demand forecasting model using a machine learning algorithm, and means for designing an inventory allocation plan. This enables rapid collection and analysis of sales information, and efficient inventory management through highly accurate demand forecasting. Furthermore, by visually representing inventory liquidity through an information terminal and using voice support technology, users can intuitively understand the information and make quick decisions.

[0736] "Route" is a term that refers to the points or paths that goods or information take to circulate.

[0737] An "information storage device" is an electronic device or system for accumulating various types of collected data and storing them in a format that makes them easily accessible when needed.

[0738] A "machine learning algorithm" is a method or computational procedure that allows a computer to learn patterns from data and perform predictions and classifications.

[0739] An "inventory allocation plan" is a strategic plan for appropriately allocating inventory of goods and materials to each location based on demand forecasts.

[0740] A "display device" is an electronic device that visually displays information and assists users in confirming and managing it.

[0741] "Users" refer to individuals who operate systems and devices, and who utilize information to perform tasks and make decisions.

[0742] "Voice assistance technology" is a technology that uses voice to give instructions and obtain information, enabling users to access information hands-free.

[0743] A "generative AI model" is an artificial intelligence model that learns from given data and prompts and is generated to address new problems.

[0744] A "prompt statement" refers to a document or command used when inputting instructions or data into an AI model.

[0745] The system that implements this application consists of three elements: a server, a terminal, and a user. Its aim is to create an efficient environment for inventory management.

[0746] The server collects sales information and stores it in an information storage device. The collected data is used to build a demand forecasting model using machine learning algorithms. Specifically, Flask is used for data communication, and TensorFlow is used to train the forecasting model. Based on the built model, the server designs an inventory allocation plan, which is then sent to terminals via information terminals.

[0747] The terminal is responsible for visually presenting the inventory allocation plan received from the server to the user. The terminal has an application built using React Native installed, enabling data visualization using Charts.js. Furthermore, it utilizes Google Cloud Speech-to-Text as a voice assistance technology, providing users with the ability to access information hands-free.

[0748] Users use their devices to check and manage inventory information and allocation plans. The decisions made by users are sent to the server as feedback. This feedback is used to improve the model in the next cycle, enhancing the accuracy of the generated AI model.

[0749] To give a specific example, if a sudden fluctuation in demand for a particular product is predicted at a logistics center, the server immediately processes that information and instructs the terminal to allocate inventory optimally. Based on these instructions, the user optimizes inventory, enabling efficient logistics without surpluses or shortages.

[0750] An example of a prompt message for a generative AI model might be: "Create a program that forecasts demand for a specific product category and displays the results in a smartphone app."

[0751] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0752] Step 1:

[0753] The server collects sales information from each distribution channel and stores it in the information storage device. In this step, the server collects sales information into a database and organizes and stores the data. The input is raw sales data from each distribution channel, and the output is organized sales data stored in the information storage device. Data normalization and duplicate removal are performed.

[0754] Step 2:

[0755] The server uses accumulated sales data to execute a machine learning algorithm and build a demand forecasting model. In this step, the model is trained using tools such as TensorFlow. The input is organized sales data stored in an information storage device, and the output is a trained model specifically designed for demand forecasting. After data preprocessing, the model performs pattern recognition and generates forecast data.

[0756] Step 3:

[0757] The server predicts the demand for each product based on the built model and develops an inventory allocation plan. Here, the server performs predictions using the trained model and calculates the optimal inventory allocation based on the results. The input is the trained demand forecasting model and newly collected sales data, and the output is the inventory allocation plan. The forecasting algorithm identifies the supply-demand gap and automatically generates the allocation plan.

[0758] Step 4:

[0759] The terminal visually presents the inventory allocation plan received from the server to the user. In this step, the terminal uses Charts.js visualization via a React Native application. The input is the inventory allocation plan sent from the server, and the output is information presented visually and audibly on the screen and through voice assistance technologies. The terminal uses a voice assistant to inform the user of important information.

[0760] Step 5:

[0761] Users review inventory allocation plans via a terminal and take actions based on on-site conditions. They also send feedback to the server to improve the accuracy of future forecasts. In this step, users perform physical inventory operations based on the plan and provide feedback on the results. Inputs are the inventory allocation plan and on-site information displayed on the terminal, and outputs are the results of the actions taken and the feedback thereto. Based on the results, inventory optimization and model accuracy improvements are performed.

[0762] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0763] This invention combines an emotion engine with an inventory management system to optimize inventory allocation while considering the user's emotional state. The system is composed of three elements: a server, a terminal, and a user.

[0764] server

[0765] The server builds a demand forecasting model using machine learning algorithms based on sales data collected from various distribution channels. In addition, the emotion engine processes emotional data obtained from user input and actions and incorporates it into the demand forecasting model. The emotion engine analyzes user text input and voice data using natural language processing and sentiment analysis algorithms.

[0766] terminal

[0767] When the terminal presents the user with an inventory allocation plan, the emotion engine displays information tailored to the user's emotional state. For example, if the user is stressed, the terminal provides concise, real-time information and applies an interface to assist in decision-making.

[0768] User

[0769] Users can view inventory data and demand forecasts through their devices to make decisions for the next steps. Sentimental data obtained from user interactions is fed back to the server, which uses this data to continuously improve its forecasting models and displayed content.

[0770] Specific example

[0771] For example, if the emotion engine detects signs such as a user spending a long time working on the inventory management screen and their input speed slowing down, the server records this as a record, and the next time a similar situation is detected, the terminal will send a simplified notification. Also, if the emotion engine detects a state of excitement when demand for a particular product suddenly increases, the server will immediately recommend high-priority inventory management actions.

[0772] In this way, the system provides a form that supports efficient decision-making for users by performing highly accurate inventory management through the collection of sales data and the analysis of sentiment.

[0773] The following describes the processing flow.

[0774] Step 1:

[0775] The server periodically collects sales data from each distribution channel and stores it in a database. This includes sales history, inventory status, and seasonal sales characteristics. The data is then processed using data cleaning algorithms to make it ready for analysis.

[0776] Step 2:

[0777] The server uses an emotion engine to analyze the user's emotional state from their operation logs and input. The emotion engine uses natural language processing and machine learning models to analyze text and voice input and recognize the user's emotional state.

[0778] Step 3:

[0779] The server combines sales data and user sentiment data to build a demand forecasting model. This model is based on machine learning algorithms and is designed to maximize forecasting accuracy by taking into account the characteristics of each distribution channel.

[0780] Step 4:

[0781] The server develops an inventory allocation plan based on predicted demand. This may involve prioritization based on sentiment data, and optimization algorithms are used to ensure that all allocations are reasonable and meet sales criteria.

[0782] Step 5:

[0783] The terminal visually presents the inventory allocation plan received from the server to the user. It incorporates the results of the emotion engine to provide information that reassures the user. For example, if the user is experiencing high stress levels, the notification will offer an intuitive and simple action.

[0784] Step 6:

[0785] Users use their devices to review the proposed inventory allocation plan and make adjustments as needed. Users also send feedback from their devices to the server, providing information, particularly regarding the effectiveness of sentiment-based responses.

[0786] Step 7:

[0787] The server analyzes user feedback and actual sales results, and based on this, continuously improves its demand forecasting model and sentiment analysis algorithm. As a result, the system's accuracy and efficiency improve over time.

[0788] (Example 2)

[0789] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0790] Conventional inventory management systems rely solely on historical sales data for demand forecasting, resulting in low accuracy in predicting actual demand fluctuations. Furthermore, they fail to consider the impact of user emotional states on inventory allocation planning decisions, potentially leading to inefficient decision-making. Therefore, a more accurate inventory allocation method that reflects user emotional states is needed.

[0791] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0792] In this invention, the server includes means for collecting sales information from each distribution channel and storing it in an information storage device, means for constructing a structure for making demand forecasts based on the information using machine learning technology, and means for processing user emotional information using an emotional analysis engine and reflecting it in the demand forecasting structure. This enables highly accurate demand forecasting and inventory allocation that takes into account the emotional state of the user.

[0793] "Distribution channels" refer to the routes and networks through which products and services are delivered from producers to consumers.

[0794] "Sales information" refers to information that includes data on product sales and sales status, and is used for forecasting future demand.

[0795] An "information storage device" includes hardware and software for storing collected data and accessing and processing it as needed.

[0796] "Machine learning techniques" is a general term for algorithms and methods that can learn patterns from large amounts of data and perform predictions and classifications.

[0797] "Structure" refers to algorithms and data flows used to process information and build models, and is a crucial element in system design.

[0798] A "sentiment analysis engine" is a software technology that uses natural language processing and speech processing to analyze human emotions and utilize them as data.

[0799] "Users" refer to individuals who operate the system, input data, and make decisions.

[0800] "Emotional information" refers to data that indicates the emotional state of a user, analyzed based on their actions and inputs.

[0801] "Demand forecasting" refers to the process of predicting future demand for a product based on past sales information and current data.

[0802] This invention is a new type of system that combines an inventory management system and an emotion engine, and its core is based on data exchange between a server, a terminal, and a user.

[0803] server

[0804] The server primarily collects sales information from various distribution channels and stores it in a database, which serves as the information storage device. This process involves periodically executing data collection scripts and retrieving sales information via APIs and other data acquisition methods. The collected information is used to build demand forecasting models using machine learning techniques. Specifically, random forest and linear regression models are trained using Python's Scikit-learn. Furthermore, the server utilizes a sentiment analysis engine to analyze text input and audio data from users and extract sentiment information. Natural language processing libraries such as NLTK and Transformers are used for this process. The analyzed sentiment information is then incorporated as new features into the demand forecasting model, resulting in more accurate predictions.

[0805] terminal

[0806] When the terminal presents the inventory allocation plan transmitted from the server to the user, the information display is adjusted based on the results of the sentiment analysis engine. For example, if the user's stress level is detected, the information is provided in a simplified form in real time, and an interface is applied to support necessary decision-making. This makes it easier for the user to understand and make quick decisions.

[0807] User

[0808] Users access inventory data and demand forecasts through their terminals and make decisions for the next steps based on that information. Sentimental information and feedback data obtained through user interaction are fed back to the server and used for continuous improvement of the demand forecasting model and displayed content. In this way, the entire system forms a dynamic improvement cycle that takes user emotions into account, enhancing the user experience.

[0809] Specific example

[0810] For example, if the emotion engine detects a user's level of interest by frequently checking the inventory status of a particular product, it will prioritize displaying inventory data for that product the next time the user checks. In addition, if the analysis reveals that the user is experiencing fatigue due to slow input speed, the device can simplify the information and display only the most important data.

[0811] Examples of prompts for generative AI models

[0812] "Based on this week's sales data and user sentiment trends, please propose an inventory allocation plan for next week."

[0813] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0814] Step 1:

[0815] The server collects sales information from each distribution channel. The input is raw sales data obtained via APIs and other data acquisition methods. The server organizes this data and stores it in an information storage device. Data cleaning is performed to remove inaccurate data before it is stored in the database. The output is a clean set of sales information.

[0816] Step 2:

[0817] The server builds a demand forecasting model using sales information. The input is the set of sales information obtained in the previous step. The server uses Scikit-learn, a machine learning technique, to train the model by applying algorithms such as random forest and linear regression. During this process, distribution and trend analysis is performed, and a model for predicting future demand is output.

[0818] Step 3:

[0819] The server performs user sentiment analysis. The input is text or voice data from the user. The server analyzes this data using a natural language processing library and extracts the user's sentiment information. The sentiment analysis engine identifies emotional states such as positive, negative, or neutral, and provides sentiment information as output.

[0820] Step 4:

[0821] The server incorporates the acquired sentiment information into the demand forecasting model. The inputs are the demand forecasting model and the sentiment information. The server adds the sentiment information as a feature of the model and readjusts it. This readjustment enables highly accurate forecasts that take emotional states into account. The output is the demand forecasting model with sentiment reflected.

[0822] Step 5:

[0823] The terminal develops and presents an inventory allocation plan to the user based on a demand forecasting model that incorporates sentiment from the server. The inputs are the demand forecasting model and inventory data. The terminal analyzes the information and provides the user with an interface for visualization. During this process, the displayed content is adjusted according to the user's emotional state, and the adjusted inventory allocation plan is presented to the user as output.

[0824] Step 6:

[0825] The user reviews the presented inventory allocation plan and makes adjustments as needed. The input is the inventory allocation plan presented by the terminal. User interaction through the interface generates feedback information regarding the plan. The output is the information sent to the server as user feedback data.

[0826] Step 7:

[0827] The server improves the model based on user feedback. The inputs are feedback data and the current demand forecasting model. The server analyzes the feedback and tunes the model parameters as needed. This improves forecasting accuracy, resulting in an improved demand forecasting model as output.

[0828] (Application Example 2)

[0829] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0830] Traditional inventory management systems perform demand forecasting and inventory allocation based on sales data, but they do not take into account the emotional state of users, and therefore cannot be said to have fully achieved optimization of inventory allocation. In particular, when users are experiencing stress or anxiety, the way inventory information is provided may not be appropriate. In such situations, the efficiency of decision-making decreases, and as a result, the accuracy of inventory management is compromised.

[0831] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0832] In this invention, the server includes means for collecting sales data from each distribution channel and storing it in a database, means for constructing a demand forecasting model using a machine learning algorithm based on the data, and means for influencing the forecasted demand based on the user's emotional state. This makes it possible to optimize inventory allocation while taking into account the user's emotional state.

[0833] "Sales data" refers to records of how products were sold through each distribution channel.

[0834] A "database" is a collection of data built to efficiently store, manage, and retrieve collected information.

[0835] A "machine learning algorithm" is a computational method used to build predictive models that improve themselves using collected data.

[0836] "User emotional state" refers to information about the psychological state or changes a user exhibits when operating a system.

[0837] An "inventory allocation plan" is a strategy for efficiently distributing goods to each location based on predicted demand.

[0838] A "display device" is a device used to visually present information to a user.

[0839] "Feedback" is the process of returning actual performance data to the system and using it to improve models and plans.

[0840] The system that implements this application example includes a series of processes to optimize inventory management by taking into account the user's emotional state. A detailed explanation of how the system works is provided below.

[0841] The server collects sales data from each distribution channel and stores it in a database. This prepares the foundational data for demand forecasting. As a machine learning algorithm, "scikit-learn" is used to analyze the sales data and build a demand forecasting model. Furthermore, the sentiment engine extracts sentiment data from user input and actions, and analyzes it using natural language processing libraries such as "NLTK" and "TextBlob". This sentiment data is then reflected in the demand forecasting model.

[0842] The device displays information tailored to the user's emotional state, and if the user is feeling stressed, it presents information in a concise and concise manner. This allows the user to make quick and accurate decisions.

[0843] Based on the inventory data and forecast information displayed on the device, users decide what action to take next. Sentiment data obtained from user interactions is constantly fed back to the server and used to improve the accuracy of the model and optimize the displayed content.

[0844] A concrete example would be a scenario where an employee working at a logistics center gives a voice command via their smartphone, saying, "Tell me the latest demand forecast." In this case, if the system senses anxiety from the voice, it will concisely display only the most important information visually. An example of a prompt message would be, "Infer anxiety from the voice and display inventory information accordingly."

[0845] This system streamlines the inventory management process at logistics centers and reduces the psychological burden on users.

[0846] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0847] Step 1:

[0848] The server collects sales data from each distribution channel and stores it in a database. This sales data contains historical information about when and through which channels products were sold. Sales logs from each store and distributor are provided as input, and this data is organized and stored in the database as output. Specifically, the server performs data insertion into the database and consistency checks.

[0849] Step 2:

[0850] The server uses collected sales data to build a demand forecasting model based on a machine learning algorithm. Sales performance and distribution channel characteristics are provided as input data, and the demand forecasting model is generated as output. This process utilizes "scikit-learn" for data preprocessing, algorithm selection, model training, and evaluation of prediction accuracy.

[0851] Step 3:

[0852] The server analyzes voice or text data obtained from user input and actions using natural language processing tools such as "NLTK" and "TextBlob" to evaluate the user's emotional state. Voice data or text chat is provided as input, and the user's emotional state is obtained as output. Specific operations include text conversion, sentiment analysis, and identification of emotional tendencies.

[0853] Step 4:

[0854] The server updates the demand forecasting model based on the sentiment data obtained from the sentiment engine. Sentiment data is input, and a new demand forecast reflecting the sentiment is created as output. In this process, the model is retrained to incorporate the sentiment data.

[0855] Step 5:

[0856] The terminal receives the inventory allocation plan provided by the server and displays it to the user. Here, the information is simplified according to the user's emotional state. Inventory allocation information is received as input, and a visual display tailored to the user's emotions is produced as output. Specifically, filtering and highlighting of the displayed content are performed.

[0857] Step 6:

[0858] The user makes decisions based on the presented inventory allocation plan and feeds the results back to the server. The inventory allocation plan is displayed as input, and feedback data is generated as output. Specifically, the actions decided by the user are sent to the server and used for future model improvements of the system.

[0859] The specific processing unit 290 transmits the result of the specific processing 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.

[0860] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. 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. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0861] 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.

[0862] 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.

[0863] 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.

[0864] 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.

[0865] 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.

[0866] 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.

[0867] 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."

[0868] 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.

[0869] 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.

[0870] 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.

[0871] 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.

[0872] 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.

[0873] 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.

[0874] 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.

[0875] 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.

[0876] 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.

[0877] 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.

[0878] 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.

[0879] 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.

[0880] The following is further disclosed regarding the embodiments described above.

[0881] (Claim 1)

[0882] A means of collecting sales data from each distribution channel and storing it in a database,

[0883] A means for constructing a model for demand forecasting based on the aforementioned data using a machine learning algorithm,

[0884] A means of formulating an inventory allocation plan based on predicted demand,

[0885] Means for providing the aforementioned allocation plan to the user using a display device,

[0886] A system that includes means for collecting actual sales performance as feedback and improving the aforementioned model.

[0887] (Claim 2)

[0888] The system according to claim 1, comprising means for improving forecasting accuracy by taking into account characteristic data for each distribution channel when performing the aforementioned demand forecast.

[0889] (Claim 3)

[0890] The system according to claim 1, comprising means for using an optimization algorithm that takes into account warehouse capacity limitations and transportation costs when formulating the inventory allocation plan.

[0891] "Example 1"

[0892] (Claim 1)

[0893] A means for collecting sales information at each distribution channel and storing it in an information storage device,

[0894] A means for constructing a forecasting model for demand forecasting based on the aforementioned information using machine learning techniques,

[0895] Means for establishing an inventory allocation plan based on predicted demand,

[0896] A means for presenting the aforementioned allocation plan to the user using a display device,

[0897] A means for collecting actual sales performance as feedback and improving the aforementioned prediction model,

[0898] A system including means for analyzing the aforementioned feedback information and improving the accuracy of a predictive model using a generated AI model.

[0899] (Claim 2)

[0900] The system according to claim 1, comprising means for improving the accuracy of the forecast by taking into account characteristic information for each distribution channel and further using a generative AI model when performing the aforementioned demand forecast.

[0901] (Claim 3)

[0902] The system according to claim 1, further comprising means for using an optimization method that takes into account the capacity limitations of storage facilities and logistics costs when formulating the aforementioned inventory allocation plan.

[0903] "Application Example 1"

[0904] (Claim 1)

[0905] A means of collecting sales information from each route and storing it in an information storage device,

[0906] A means for constructing a model for demand forecasting based on the aforementioned information using a machine learning algorithm,

[0907] Means for designing inventory allocation plans based on predicted demand,

[0908] A means for presenting the aforementioned allocation plan to the user using a display device,

[0909] A means of collecting actual sales results as feedback and improving the aforementioned model,

[0910] When implementing the aforementioned inventory allocation plan, a means is provided to visually represent inventory liquidity via an information terminal and to facilitate information confirmation by users using voice support technology.

[0911] A means to improve prediction accuracy by using prompt sentences through a generative AI model,

[0912] A system that includes this.

[0913] (Claim 2)

[0914] The system according to claim 1, comprising means for improving forecasting accuracy by taking into account characteristic information for each route when performing the aforementioned demand forecast.

[0915] (Claim 3)

[0916] The system according to claim 1, comprising means for using an optimization algorithm that takes into account the capacity limitations of storage facilities and transfer costs when designing the inventory allocation plan.

[0917] "Example 2 of combining an emotion engine"

[0918] (Claim 1)

[0919] A means for collecting sales information from each distribution channel and storing it in an information storage device,

[0920] A means for constructing a structure for performing demand forecasting based on the aforementioned information using machine learning technology,

[0921] A means for processing user emotional information using an emotion analysis engine and reflecting it in the demand forecasting structure,

[0922] A means for formulating an inventory allocation plan based on the results reflected above,

[0923] Means for providing the aforementioned plan to the user using a display device,

[0924] A system including means for collecting feedback information obtained from user operations and improving the structure.

[0925] (Claim 2)

[0926] The system according to claim 1, comprising means for improving forecasting accuracy by taking into account characteristic information for each sales channel and user sentiment information when performing the aforementioned demand forecast.

[0927] (Claim 3)

[0928] The system according to claim 1, comprising means for using optimization techniques that take into account the capacity limitations of storage facilities and transportation costs when formulating the aforementioned inventory distribution plan.

[0929] "Application example 2 when combining with an emotional engine"

[0930] (Claim 1)

[0931] A means of collecting sales data from each distribution channel and storing it in a database,

[0932] A means for constructing a model for demand forecasting based on the aforementioned data using a machine learning algorithm,

[0933] Means of influencing predicted demand based on the user's emotional state,

[0934] A means of formulating an inventory allocation plan based on predicted demand,

[0935] Means for providing the aforementioned allocation plan to the user using a display device,

[0936] A system that includes means for collecting actual sales performance as feedback and improving the aforementioned model.

[0937] (Claim 2)

[0938] The system according to claim 1, comprising means for improving forecasting accuracy by taking into account characteristic data for each distribution channel when performing the aforementioned demand forecast.

[0939] (Claim 3)

[0940] The system according to claim 1, comprising means for using an optimization algorithm that takes into account warehouse capacity limitations and transportation costs when formulating the inventory allocation plan. [Explanation of Symbols]

[0941] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting sales information from each route and storing it in an information storage device, A means for constructing a model for demand forecasting based on the aforementioned information using a machine learning algorithm, Means for designing inventory allocation plans based on predicted demand, A means for presenting the aforementioned allocation plan to the user using a display device, A means of collecting actual sales results as feedback and improving the aforementioned model, When implementing the aforementioned inventory allocation plan, a means is provided to visually represent inventory liquidity via an information terminal and to facilitate information confirmation by users using voice support technology. A means to improve prediction accuracy by using prompt sentences through a generative AI model, A system that includes this.

2. The system according to claim 1, further comprising means for improving forecasting accuracy by taking into account characteristic information for each route when performing the aforementioned demand forecast.

3. The system according to claim 1, comprising means for using an optimization algorithm that takes into account the capacity limitations of storage facilities and transportation costs when designing the inventory allocation plan.

Images