system

Abstract

We provide the system. [Solution] Means of obtaining sales history information, A means of predicting demand based on acquired sales history information, A means for generating an optimal ordering plan based on predicted demand, A means for displaying the generated order plan, A system that includes this.

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, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, 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] In the retail of food products, due to purchasing and promotion relying on experience and intuition, food waste and loss have become a serious problem. Such problems are caused by different operating policies and local rules for each store, making it difficult to introduce optimal inventory management and sales promotion measures. Furthermore, existing systems are insufficient in real-time demand forecasting and automation of inventory management, and there is a need to improve operational efficiency. Therefore, a new method for reducing food loss and improving store operations is required.

Means for Solving the Problems

[0005] This invention provides a system that acquires sales history information and predicts demand based on that information. Specifically, it includes means for generating an optimal ordering plan based on the predicted demand and displaying that plan on a terminal in the store. Furthermore, it aims to reduce food waste by identifying products with a high risk of being discarded based on the demand forecast and suggesting effective sales promotion measures for these products. In addition, it provides a system that improves the accuracy of demand forecasting by evaluating the difference between actual sales data and forecast data and retraining the AI ​​model. This makes it possible to support data-driven store operations and achieve increased efficiency in management work and a reduction in food waste.

[0006] "Sales history information" refers to past sales data for each product, including detailed records such as product name, quantity sold, and date and time of sale.

[0007] "Methods for predicting demand" refers to systems that use AI models or algorithms to estimate future sales volumes based on sales history information.

[0008] "Means for generating order plans" refers to the process of calculating the optimal order quantity and timing for goods based on predicted demand, as well as the software or system that performs this process.

[0009] "Products at high risk of disposal" refer to products that are nearing their expiration date and are likely to be discarded if not sold.

[0010] "Means of proposing sales promotion measures" refers to a system for planning and presenting promotional activities that appeal to consumers, such as discounts or recipe suggestions, in order to promote the sale of a specific product.

[0011] "Methods for retraining AI models" refer to the process of analyzing the differences between actual sales data and predicted data, and then updating the parameters of the AI ​​model based on that data to improve the model's accuracy. [Brief explanation of the drawing]

[0012] [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] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0013] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, a processor with a reference number (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.

[0016] In the following embodiments, a RAM (Random Access Memory) with a reference number is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, a storage with a reference number is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0018] 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).

[0019] 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."

[0020] [First Embodiment]

[0021] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0022] 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.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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".

[0033] The following system operation is conceivable as an embodiment of the present invention. This system predicts demand based on sales history information, enables the generation of order plans, and facilitates timely sales promotion.

[0034] The server periodically retrieves sales history information from each store and uses this data to perform demand forecasting with an AI model. For example, in store A, monthly sales history is sent to the server, where the data is analyzed by an AI algorithm to forecast demand for the following month.

[0035] The server generates an optimal ordering plan based on predicted demand. This ordering plan details purchase quantities and timings, and is optimized to minimize excess inventory and stockouts. The generated ordering plan is sent to each store's terminal, allowing store staff to use it to order their next products.

[0036] The terminals are connected to digital signage within the store and display promotional information, including products at high risk of being discarded. For example, a terminal in store B suggests recipes using ingredients nearing their expiration date and offers discounts on those products to appeal to customers. Users can then use this information to increase sales of their inventory.

[0037] The system works by having the server receive feedback on the difference between actual sales results and forecast data, and then periodically retraining the AI ​​model to improve its accuracy. At store C, information on the difference between forecasts and actual sales results is sent to the server and used to adjust the AI ​​model. In this way, the system is continuously improved, enabling more accurate demand forecasting.

[0038] This system enables store operators to make data-driven decisions, leading to more efficient inventory management and reduced food waste.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The server retrieves sales history information from each store. The server works in conjunction with the stores' POS systems and automatically imports the most recent sales data into the database on a daily basis.

[0042] Step 2:

[0043] The server preprocesses the collected sales history information. The server identifies missing or outlier data, and then fills in and corrects them to generate a dataset suitable for the AI ​​model.

[0044] Step 3:

[0045] The server performs demand forecasting using an AI model. The server uses a pre-trained time-series forecasting model to predict future sales figures for each product.

[0046] Step 4:

[0047] The server generates an optimal ordering plan based on the demand forecast. Following the demand forecast, and taking inventory levels and lead times into consideration, the server calculates the quantity and timing of orders to be placed at the lowest possible cost.

[0048] Step 5:

[0049] The server sends the order plan to the terminals in each store. The terminals receive this order plan and prepare it for display in an easy-to-read format for the person in charge.

[0050] Step 6:

[0051] The terminal displays information about products at high risk of being discarded on digital signage. The terminal also displays recipes and discount information for identified products to appeal to customers.

[0052] Step 7:

[0053] Users (store staff) plan product orders and sales promotions based on information from the terminal. Users confirm the suggested quantities on the terminal, place actual orders, and implement promotional campaigns.

[0054] Step 8:

[0055] The server collects actual sales data and retrains the AI ​​model. The server compares the predicted data with the actual data and updates the model parameters to improve the accuracy of the AI ​​model.

[0056] (Example 1)

[0057] 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."

[0058] In today's distribution industry, proper inventory management and improved demand forecasting accuracy are crucial for solving problems such as excess inventory and stockouts. However, existing systems struggle to generate precise demand forecasts and optimal ordering plans that adapt to demand fluctuations. Furthermore, sales promotion measures are not adjusted in real time, resulting in inefficient sales of products with a high risk of spoilage. Solving these challenges is essential.

[0059] 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.

[0060] In this invention, the server includes means for collecting and pre-processing sales history information, means for predicting demand using an AI model generated based on the pre-processed sales history information, and means for generating an order plan using an optimization algorithm based on the predicted demand. This enables precise demand forecasting in response to demand fluctuations and the generation of an optimal order plan.

[0061] "Sales history information" refers to records of product sales at each store, including data such as product name, quantity sold, and date and time of sale.

[0062] A "generative AI model" is a system that uses machine learning algorithms to analyze data, identify patterns, and predict future demand.

[0063] "Preprocessing" refers to a series of processes, such as data organization, outlier removal, and missing value handling, that transform raw data into a format that can be easily handled by AI models.

[0064] "Demand forecasting" is the process of estimating future demand for a product based on past sales data and related information.

[0065] An "optimization algorithm" is a method for minimizing or maximizing resource utilization to achieve a specific objective, and in this context, it is used to optimize ordering plans in inventory management.

[0066] An "ordering plan" is a detailed plan that specifies the quantity and timing of purchases of goods in accordance with predicted demand, with the aim of minimizing the risk of excess inventory or stockouts.

[0067] "Sales promotion measures" are strategies and activities undertaken to increase product sales, aiming to communicate the appeal of a product to consumers and stimulate their desire to purchase.

[0068] "Feedback" refers to information obtained from actual sales results, and analyzing this data is crucial for improving the prediction accuracy of AI models.

[0069] This invention is a system that functions around three main elements: a server, a terminal, and a user. The server first collects sales history information from each store. This information includes details such as product name, quantity sold, and date and time of sale, and is stored in a database. The collected data is preprocessed and prepared in a format that is easy for the generated AI model to handle.

[0070] Next, the server uses a generative AI model to forecast demand. This AI model is designed based on machine learning algorithms and analyzes past sales data to estimate future demand. This provides a foundation for minimizing supply-demand mismatches. For example, it can make forecasts that take into account demand fluctuation patterns in line with specific seasons or events.

[0071] Furthermore, the server generates an order plan using an optimization algorithm based on the demand forecast results. This plan includes specific product quantities and order timings, and is designed to avoid excess inventory and stockouts while considering inventory costs. The order plan is sent to terminals, where store staff can use it to manage inventory efficiently.

[0072] The terminals are connected to digital signage used within the store, displaying sales promotion information in real time. This is particularly used to promote products at high risk of spoilage, aiming to increase sales by conveying the appeal of the products to customers. For example, for products nearing their expiration date, related recipes and discount information are displayed to stimulate purchasing intent.

[0073] Users provide actual sales results to the server, and this data is used as feedback to retrain the AI ​​model. This continuously improves the accuracy of the AI ​​model, enabling more accurate demand forecasting. Through this process, the system can automatically learn and evolve.

[0074] As a specific example, store D uses the prompt message, "Based on sales history over the past three months, predict the demand for a specific product for the next month," to perform demand forecasting. Based on this result, an ordering plan can be created, and the results can be used for promotions on terminals.

[0075] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0076] Step 1:

[0077] The server periodically collects sales history information from each store. This information is input as a dataset containing details such as product name, sales quantity, and sales date and time. The server processes this data by handling outliers and missing values, transforming it into well-formed data to generate preprocessed data suitable for input to the AI ​​model.

[0078] Step 2:

[0079] The server inputs pre-processed data into a generating AI model and performs demand forecasting. This process involves data calculations that recognize past sales patterns and use time series analysis to predict future consumption trends. The output is a demand forecast value for each product.

[0080] Step 3:

[0081] The server generates an order plan using an optimization algorithm based on predicted demand values. The input is the demand forecast, and the algorithm calculates the order quantity and purchase timing for each product, taking into account order lead time, inventory costs, and stockout risk. The output is the optimal order plan.

[0082] Step 4:

[0083] The server sends the generated order plan to the terminals in each store. Based on the received order plan, the terminals display appropriate sales promotion information on digital signage. For example, for products nearing their expiration date, the signage displays relevant recipes and discount information to encourage customer purchases.

[0084] Step 5:

[0085] Users provide actual sales results to the server. This feedback data, including sales quantity and time information, is used as input for analysis on the server. The data analysis evaluates the difference from the predicted results and manages it as output data to facilitate the retraining of the AI ​​model.

[0086] Step 6:

[0087] The server periodically retrains the generated AI model using feedback data. This involves using difference information as input data, updating the model, and improving its accuracy. The expected output is improved model accuracy in the next demand forecast.

[0088] (Application Example 1)

[0089] 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."

[0090] In today's distribution and retail industries, improving the accuracy of inventory management and demand forecasting is an urgent issue. The risk of waste due to excess inventory and lost sales opportunities due to stockouts directly impact operational efficiency and profits, thus increasing the need for systems that comprehensively address these issues. Furthermore, there is a demand for mechanisms that enable store staff to efficiently obtain information and take action.

[0091] 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.

[0092] In this invention, the server includes means for acquiring sales history information, means for predicting demand based on the acquired sales history information, means for generating an optimal order plan based on the predicted demand, means for displaying the generated order plan, and means for providing the generated order plan and sales promotion information to store users via customer terminals. This enables efficient inventory management and accurate demand forecasting, thereby optimizing store operations.

[0093] "Sales history information" refers to information that records past sales data at each sales location, and is the basis for demand forecasting and inventory management.

[0094] "Means of predicting demand" refer to technical elements that analyze acquired sales history information to estimate future sales volumes and customer purchasing trends.

[0095] An "ordering plan" is a set of specific guidelines and plans for ordering the necessary goods in the appropriate quantities within the appropriate timeframe, based on demand forecasts.

[0096] "Means of display" refers to equipment or devices that output generated information in a visually recognizable format, and plays a role in conveying important information to store staff and customers.

[0097] A "customer terminal" is an information device that can be operated by customers visiting a store, and is used to receive necessary information and instructions.

[0098] "Sales promotion information" refers to information designed to encourage the purchase of a specific product, and mainly includes discounts, campaigns, and product benefits.

[0099] An "electronic display device" is a device that displays information in digital format and is used to visually present information within a store.

[0100] A "generative algorithm" is a series of computational procedures and methods used in information processing to analyze data and derive results.

[0101] A "prompt statement" is an input statement used to provide specific instructions or information to a generation algorithm in order to obtain the desired output.

[0102] As an embodiment for carrying out the present invention, the system has the following configuration. The server periodically acquires sales history information from each sales location. The sales history information is data on what products the store has sold and in what quantities in the past, and is basic data used for demand forecasting.

[0103] The server feeds the acquired sales history information into an AI model for demand forecasting. This AI model is built using machine learning libraries such as TENSORFLOW® and PyTorch, and is continuously retrained to improve prediction accuracy. Based on the predicted demand, the server generates an optimal ordering plan. The ordering plan indicates which products to order, in what quantities, and at what time of year the orders should be placed.

[0104] The generated order plans and sales promotion information are visually displayed on electronic display devices within the store or provided to store users via customer terminals. This allows sales promotion information to be conveyed to customers in real time, making it possible to encourage the purchase of products that are at high risk of being discarded.

[0105] Furthermore, the server collects this discrepancy information as feedback to evaluate the differences between actual sales data and forecast data, and to continuously improve the AI ​​model. For example, the server generates a prompt message such as "Which products are likely to see increased sales in the near future?" and adjusts the algorithm to achieve more accurate predictions.

[0106] This system allows users to easily obtain highly accurate demand forecasts based on sales history information and optimal ordering plans accordingly, thereby improving the efficiency of inventory management and sales promotion.

[0107] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0108] Step 1:

[0109] The server retrieves sales history information from each sales location. This sales history information is a digital record of past sales data at each store. Input data includes product ID, sales date, and sales quantity. The retrieved data is stored in the server's database.

[0110] Step 2:

[0111] The server uses sales history data to perform demand forecasting using an AI model. This model is built using TensorFlow and performs time series analysis. Sales history data is used as input, and the output generates predicted sales volume for the next specified period. The server runs the AI ​​model and obtains the forecast results.

[0112] Step 3:

[0113] The server generates an optimal ordering plan based on predicted demand. The input is the results of the demand forecast, and the output is a plan that includes the order quantity and timing for each product. This ordering plan also takes into account inventory levels and past ordering history.

[0114] Step 4:

[0115] Order plans and sales promotion information are transmitted from the server to electronic display devices and customer terminals in stores. Input data includes order plan information and sales promotion strategies, and output generates visual information to be provided to store staff and customers.

[0116] Step 5:

[0117] Users manage inventory and conduct sales promotion activities based on the displayed information. By operating the terminal and accessing sales promotion information, users can process inventory items and promote products nearing their expiration date.

[0118] Step 6:

[0119] The difference between actual sales data and forecast data is fed back to the server. This feedback allows the AI ​​model to be retrained, improving the accuracy of demand forecasting. The difference information is then used as input for further data processing, updating the AI ​​model to aim for more accurate forecasts.

[0120] Step 7:

[0121] The server continuously generates prompts to adjust the algorithm and tune the model. For example, it generates a prompt such as "Which products are likely to see increased sales next week?" to adjust the model and maximize operational effectiveness.

[0122] 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.

[0123] One possible embodiment of the present invention is the operation of a system incorporating an emotion engine. This system provides demand forecasting and optimized ordering plans based on sales history information, as well as a function to recognize user emotions and adjust sales promotion measures accordingly.

[0124] The server first retrieves sales history information from each store and uses an AI model to forecast demand based on that data. Based on the forecast results, the server generates an optimal ordering plan and sends this information to the store's terminal. For example, at store A, this month's sales history is sent to the server, and demand for the following month is forecasted.

[0125] The terminal displays received order plans to store staff and uses an emotion engine to recognize customers' emotions in real time. This emotion information is used, for example, for sales promotion through in-store digital signage. As a specific example, in store B, a camera scans the faces of customers who come in, and the emotion engine recognizes emotions such as joy or surprise from their facial expressions.

[0126] Based on the emotions it recognizes, the terminal adjusts sales promotion strategies. For example, it can provide discount information on products the customer has shown interest in, or suggest products that match a specific emotion. Store staff then use the suggestions from the terminal to optimize on-site customer service and product placement.

[0127] Furthermore, the server continuously collects actual sales results and forecast data, which is used to retrain the AI ​​model. The server is designed to improve the accuracy of demand forecasting and quantitatively improve the emotion engine algorithm. At store C, sales results and customer emotion reactions are sent to the server and used to improve forecast accuracy and adapt sales strategies for future events.

[0128] In this way, systems incorporating an emotion engine enable data-driven demand forecasting and real-time customer response, supporting store operations more effectively and efficiently.

[0129] The following describes the processing flow.

[0130] Step 1:

[0131] The server periodically collects sales history information from each store. The server integrates this data and stores it in a database. In this process, information such as the date of sale, product name, and quantity sold is organized.

[0132] Step 2:

[0133] The server uses an AI model to perform demand forecasting based on collected sales history information. The generated forecasts indicate the demand for each product over a certain period in the future, taking seasonality and trends into account.

[0134] Step 3:

[0135] The server generates an optimal ordering plan based on demand forecasts. The server considers product inventory levels and lead times to determine the necessary products and their order quantities. The ordering plan also includes adjustments to prevent excess inventory.

[0136] Step 4:

[0137] The server sends the order plan to terminals in each store. The terminals display this plan in a format that is easy for store staff to use, supporting quick confirmation and decision-making.

[0138] Step 5:

[0139] The device recognizes customers' emotions in real time through an emotion engine. The device acquires data from cameras and sensors installed in the store and analyzes emotions from the user's facial expressions and voice.

[0140] Step 6:

[0141] The terminal displays customized sales promotion strategies based on the customer's recognized emotional information. For example, if it determines that a customer is interested, it will display promotional content for related products on the digital signage.

[0142] Step 7:

[0143] Users (store staff) adjust sales strategies based on information provided by the terminal. Users utilize suggested discount information and sales promotion measures to optimize direct customer service and product placement.

[0144] Step 8:

[0145] The server collects the differences between actual sales data and the predictions used, and retrains the AI ​​model and emotion engine. The server analyzes this data and forms a feedback loop to improve the accuracy of the model in the future.

[0146] (Example 2)

[0147] 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".

[0148] Traditional sales systems focused on providing demand forecasts and order plans based on sales history information, but they lacked the ability to adjust sales promotion strategies based on individual customer sentiment, limiting their ability to optimize the customer experience. Furthermore, they lacked mechanisms for continuously improving the discrepancies between sales results and forecasts, which negatively impacted the accuracy of demand forecasts.

[0149] 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.

[0150] In this invention, the server includes means for collecting sales history data, means for forecasting demand based on the collected sales history data, and means for generating an optimized order plan based on the forecasted demand. This makes it possible to adjust sales promotion measures to take into account customer sentiment, thereby improving the accuracy of demand forecasting and enhancing the customer experience.

[0151] "Sales history data" refers to information such as transaction records and inventory fluctuations from past sales activities, and includes elements such as product category, sales quantity, and sales amount.

[0152] "Means of forecasting demand" refers to technical devices or methods for calculating and estimating future demand based on collected sales history data, and includes the use of machine learning models.

[0153] "Means for generating order plans" refers to technical devices or methods for creating plans to efficiently replenish goods based on predicted demand information.

[0154] A "device" refers to hardware used to visually provide generated information to a user, and includes display devices and mobile terminals.

[0155] "Means of recognizing customer emotions" refers to technologies that analyze a customer's facial expressions and actions to identify their emotional state, and includes combining cameras with emotion recognition software.

[0156] "Means of adjusting sales promotion strategies" refer to technical devices and methods that utilize customer sentiment information to formulate and execute the most appropriate sales activities and promotions at any given time.

[0157] An "artificial intelligence model" refers to a group of algorithms or programs that learn from large amounts of data and perform predictions and classifications, and includes the use of machine learning and deep learning technologies.

[0158] This invention relates to a system that integrates demand forecasting using sales history data with sales promotion that takes customer sentiment into consideration. This system, consisting of a server and terminals, streamlines sales activities and improves the customer experience.

[0159] The server is primarily responsible for data collection and analysis. First, the server collects sales history data from each store. This data includes sales volume, sales amount, and product category, and is automatically extracted from POS systems and inventory management systems. Using this data, the server predicts future demand using an AI model. Machine learning frameworks such as TensorFlow and PyTorch can be used for the AI ​​model. Based on the prediction results, the server generates an optimized ordering plan and sends it to the store's terminal.

[0160] The terminal visualizes received order plans for store staff. It also incorporates a camera and emotion recognition software, allowing it to scan customers' faces and recognize their emotions in real time. Emotion recognition software used includes tools like Amazon Rekognition, which identify emotions such as "joy" and "surprise" from customer facial expressions. Based on the recognized emotions, the terminal displays appropriate sales promotions on in-store displays and pop-ups. For example, if a customer shows a surprised expression, it can display discount information for that product.

[0161] Store staff, as users of the system, can use the information provided by the terminal to optimize customer service and product placement, thereby improving their actual sales activities. This enables data-driven decision-making, leading to more effective inventory management and increased customer satisfaction.

[0162] Furthermore, the server continuously monitors the discrepancies between sales results and forecast data, and uses this information to retrain the AI ​​model. This improves the accuracy of demand forecasting and enhances the effectiveness of promotions based on customer sentiment prediction data. Overall, the system aims to achieve both efficient sales management and high customer satisfaction.

[0163] As a concrete example, a prompt message for a generative AI model might be, "Based on customer sentiment data, please suggest the most effective discount strategy."

[0164] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0165] Step 1:

[0166] The server collects sales history data from each store. This data input includes information from POS systems and inventory management systems. The server analyzes this data and generates output that organizes and aggregates information such as sales volume, sales amount, and product category.

[0167] Step 2:

[0168] The server inputs collected sales history data into an AI model to perform demand forecasting. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to perform data calculations that estimate future demand based on patterns in past data. As a result, it outputs demand forecast data for the following month.

[0169] Step 3:

[0170] The server generates an optimized ordering plan based on demand forecast data. Here, demand data is used as input, and a supply chain management algorithm is applied to process the data to determine the quantity and timing of product replenishment. The output is a specific ordering schedule for each store.

[0171] Step 4:

[0172] The server sends the generated order plan to the store's terminal. The terminal receives the order plan from the server as input and displays it on the screen. This output includes a summary of the received order information and notification messages.

[0173] Step 5:

[0174] The terminal uses in-store cameras to scan customers' faces and recognize their emotions in real time. The input includes camera footage, which is then analyzed by emotion recognition APIs such as Amazon Rekognition. As a result, the recognized customer's emotion data is output.

[0175] Step 6:

[0176] The device adjusts sales promotion strategies based on recognized emotional data. Specifically, it takes customer emotions such as joy or surprise as input and displays appropriate discount information and promotional content on the screen. The output is the adjusted sales promotion information.

[0177] Step 7:

[0178] The server monitors the difference between actual sales data and forecast data. Using this difference data as input, the server performs data calculations to improve the accuracy of demand forecasting by retraining the AI ​​model. The output is the updated AI model parameters.

[0179] (Application Example 2)

[0180] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0181] Traditional store operations were limited to simple demand forecasting and ordering plans based on sales history data, and did not incorporate sales promotion strategies that considered consumers' real-time emotions. As a result, there were challenges in improving the customer experience and maximizing the effectiveness of sales promotions. Furthermore, the lack of continuous improvement of AI models based on sales results limited the accuracy of demand forecasting.

[0182] 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.

[0183] In this invention, the server includes means for acquiring sales history information, means for recognizing a consumer's facial expression and estimating their emotions, and means for generating sales promotion information based on the estimated emotions. This enables the implementation of sales promotion measures that reflect consumer emotions in real time, and continuous improvement of the AI ​​model based on sales results.

[0184] "Sales history information" refers to record data about past sales activities at each store, including information such as product name, quantity, date and time, and price.

[0185] "Demand forecasting" is a technique that uses statistical methods and AI models to estimate future sales volume and revenue based on past sales history information.

[0186] An "optimal ordering plan" is a strategy, including a list of products and quantities to be purchased, developed in response to predicted demand to minimize inventory costs and prevent stockouts.

[0187] "Methods for recognizing consumer facial expressions and estimating emotions" refers to technologies that use image processing techniques and machine learning algorithms to analyze emotions from facial images of consumers captured by a camera.

[0188] "Means of generating sales promotion information" refers to methods for determining and presenting marketing measures such as product discounts and purchase recommendations based on consumer sentiment and demand forecasts.

[0189] "Retraining an AI model" is the process of retraining an existing artificial intelligence model using actual sales data or newly acquired information to improve its prediction accuracy.

[0190] To implement this invention, the following system configuration is adopted. The server first acquires sales history information from each store and performs demand forecasting using an AI model. The main software used includes Python and TensorFlow. This analyzes the sales history data and predicts the number of products needed for the next period. Based on the forecast results, the server generates an optimal ordering plan and sends it to the terminals of each store.

[0191] The terminal not only displays received order plans to store staff, but also uses a camera to scan consumers' facial expressions in real time and recognize their emotions. Using a combination of OpenCV and TensorFlow, it classifies emotions into multiple categories such as joy and surprise, and generates sales promotion information based on this classification. This generated information is then presented to consumers via digital signage and smartphone displays.

[0192] Users (in this case, store staff) can adjust sales promotion strategies based on the generated emotional information. This enables personalized discount information and product recommendations for consumers. Furthermore, by feeding back newly acquired sales data and consumer emotional responses to the server, the AI ​​model is retrained, contributing to improved prediction accuracy.

[0193] For example, if a consumer smiles when looking at a particular product in a store, the terminal can display information such as "Buy this product now and get a 10% discount." Examples of prompts used for emotion recognition include "Present discount information when a smile is detected" and "Suggest related products when an expression of surprise is detected." In this way, appropriate responses based on consumer emotions become possible, contributing to improved store operational efficiency and increased customer satisfaction.

[0194] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0195] Step 1:

[0196] The server retrieves sales history information from each store. The input is sales history data for each individual store, and the output is integrated sales data. This data is collected using a Python script and stored in a database to prepare for future demand forecasting.

[0197] Step 2:

[0198] The server uses acquired sales history information to perform demand forecasting with an AI model. The input is integrated sales data, and the output is the predicted demand value. A TensorFlow model is used to analyze the data and generate estimated demand for each store.

[0199] Step 3:

[0200] The server generates an optimal ordering plan based on predicted demand. The input is predicted demand data, and the output is ordering plan data. An algorithm is used to calculate the order quantity and create the ordering plan.

[0201] Step 4:

[0202] The terminal displays the order plan received from the server to the store staff. The input is the order plan data, and the output is the information displayed on the screen. The terminal's GUI makes the information easy for staff to understand.

[0203] Step 5:

[0204] The device captures the customer's facial expressions using a camera in the store and recognizes their emotions. The input is the camera image, and the output is emotion data. Face detection is performed using OpenCV, and emotions are estimated using a TensorFlow model.

[0205] Step 6:

[0206] The device generates sales promotion information based on estimated emotions. The input is emotion data, and the output is sales promotion information. Based on the AI-generated promotional information, it provides consumers with product discounts and suggestions tailored to their needs.

[0207] Step 7:

[0208] Users adjust customer interactions based on the generated sales promotion information. The input is sales promotion strategy information, and the output is the content presented to the customer. Store staff make specific suggestions to customers, stimulating their purchasing intent.

[0209] Step 8:

[0210] The server collects actual sales data and consumer sentiment responses, and uses this data to retrain the AI ​​model. The input is sales results data and sentiment data, and the output is an improved AI model. The model is continuously updated to improve prediction accuracy.

[0211] 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.

[0212] 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.

[0213] 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.

[0214] [Second Embodiment]

[0215] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0216] 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.

[0217] 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).

[0218] 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.

[0219] 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.

[0220] 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).

[0221] 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.

[0222] 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.

[0223] 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.

[0224] 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.

[0225] 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.

[0226] 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".

[0227] The following system operation is conceivable as an embodiment of the present invention. This system predicts demand based on sales history information, enables the generation of order plans, and facilitates timely sales promotion.

[0228] The server periodically retrieves sales history information from each store and uses this data to perform demand forecasting with an AI model. For example, in store A, monthly sales history is sent to the server, where the data is analyzed by an AI algorithm to forecast demand for the following month.

[0229] The server generates an optimal ordering plan based on predicted demand. This ordering plan details purchase quantities and timings, and is optimized to minimize excess inventory and stockouts. The generated ordering plan is sent to each store's terminal, allowing store staff to use it to order their next products.

[0230] The terminals are connected to digital signage within the store and display promotional information, including products at high risk of being discarded. For example, a terminal in store B suggests recipes using ingredients nearing their expiration date and offers discounts on those products to appeal to customers. Users can then use this information to increase sales of their inventory.

[0231] The system works by having the server receive feedback on the difference between actual sales results and forecast data, and then periodically retraining the AI ​​model to improve its accuracy. At store C, information on the difference between forecasts and actual sales results is sent to the server and used to adjust the AI ​​model. In this way, the system is continuously improved, enabling more accurate demand forecasting.

[0232] This system enables store operators to make data-driven decisions, leading to more efficient inventory management and reduced food waste.

[0233] The following describes the processing flow.

[0234] Step 1:

[0235] The server retrieves sales history information from each store. The server works in conjunction with the stores' POS systems and automatically imports the most recent sales data into the database on a daily basis.

[0236] Step 2:

[0237] The server preprocesses the collected sales history information. The server identifies missing or outlier data, and then fills in and corrects them to generate a dataset suitable for the AI ​​model.

[0238] Step 3:

[0239] The server performs demand forecasting using an AI model. The server uses a pre-trained time-series forecasting model to predict future sales figures for each product.

[0240] Step 4:

[0241] The server generates an optimal ordering plan based on the demand forecast. Following the demand forecast, and taking inventory levels and lead times into consideration, the server calculates the quantity and timing of orders to be placed at the lowest possible cost.

[0242] Step 5:

[0243] The server sends the order plan to the terminals in each store. The terminals receive this order plan and prepare it for display in an easy-to-read format for the person in charge.

[0244] Step 6:

[0245] The terminal displays information about products at high risk of being discarded on digital signage. The terminal also displays recipes and discount information for identified products to appeal to customers.

[0246] Step 7:

[0247] Users (store staff) plan product orders and sales promotions based on information from the terminal. Users confirm the suggested quantities on the terminal, place actual orders, and implement promotional campaigns.

[0248] Step 8:

[0249] The server collects actual sales data and retrains the AI ​​model. The server compares the predicted data with the actual data and updates the model parameters to improve the accuracy of the AI ​​model.

[0250] (Example 1)

[0251] 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."

[0252] In today's distribution industry, proper inventory management and improved demand forecasting accuracy are crucial for solving problems such as excess inventory and stockouts. However, existing systems struggle to generate precise demand forecasts and optimal ordering plans that adapt to demand fluctuations. Furthermore, sales promotion measures are not adjusted in real time, resulting in inefficient sales of products with a high risk of spoilage. Solving these challenges is essential.

[0253] 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.

[0254] In this invention, the server includes means for collecting and pre-processing sales history information, means for predicting demand using an AI model generated based on the pre-processed sales history information, and means for generating an order plan using an optimization algorithm based on the predicted demand. This enables precise demand forecasting in response to demand fluctuations and the generation of an optimal order plan.

[0255] "Sales history information" refers to records of product sales at each store, including data such as product name, quantity sold, and date and time of sale.

[0256] A "generative AI model" is a system that uses machine learning algorithms to analyze data, identify patterns, and predict future demand.

[0257] "Preprocessing" refers to a series of processes, such as data organization, outlier removal, and missing value handling, that transform raw data into a format that can be easily handled by AI models.

[0258] "Demand forecasting" is the process of estimating future demand for a product based on past sales data and related information.

[0259] An "optimization algorithm" is a method for minimizing or maximizing resource utilization to achieve a specific objective, and in this context, it is used to optimize ordering plans in inventory management.

[0260] An "ordering plan" is a detailed plan that specifies the quantity and timing of purchases of goods in accordance with predicted demand, with the aim of minimizing the risk of excess inventory or stockouts.

[0261] "Sales promotion measures" are strategies and activities undertaken to increase product sales, aiming to communicate the appeal of a product to consumers and stimulate their desire to purchase.

[0262] "Feedback" refers to information obtained from actual sales results, and analyzing this data is crucial for improving the prediction accuracy of AI models.

[0263] This invention is a system that functions around three main elements: a server, a terminal, and a user. The server first collects sales history information from each store. This information includes details such as product name, quantity sold, and date and time of sale, and is stored in a database. The collected data is preprocessed and prepared in a format that is easy for the generated AI model to handle.

[0264] Next, the server uses a generative AI model to forecast demand. This AI model is designed based on machine learning algorithms and analyzes past sales data to estimate future demand. This provides a foundation for minimizing supply-demand mismatches. For example, it can make forecasts that take into account demand fluctuation patterns in line with specific seasons or events.

[0265] Furthermore, the server generates an order plan using an optimization algorithm based on the demand forecast results. This plan includes specific product quantities and order timings, and is designed to avoid excess inventory and stockouts while considering inventory costs. The order plan is sent to terminals, where store staff can use it to manage inventory efficiently.

[0266] The terminals are connected to digital signage used within the store, displaying sales promotion information in real time. This is particularly used to promote products at high risk of spoilage, aiming to increase sales by conveying the appeal of the products to customers. For example, for products nearing their expiration date, related recipes and discount information are displayed to stimulate purchasing intent.

[0267] Users provide actual sales results to the server, and this data is used as feedback to retrain the AI ​​model. This continuously improves the accuracy of the AI ​​model, enabling more accurate demand forecasting. Through this process, the system can automatically learn and evolve.

[0268] As a specific example, store D uses the prompt message, "Based on sales history over the past three months, predict the demand for a specific product for the next month," to perform demand forecasting. Based on this result, an ordering plan can be created, and the results can be used for promotions on terminals.

[0269] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0270] Step 1:

[0271] The server periodically collects sales history information from each store. This information is input as a dataset containing details such as product name, sales quantity, and sales date and time. The server processes this data by handling outliers and missing values, transforming it into well-formed data to generate preprocessed data suitable for input to the AI ​​model.

[0272] Step 2:

[0273] The server inputs pre-processed data into a generating AI model and performs demand forecasting. This process involves data calculations that recognize past sales patterns and use time series analysis to predict future consumption trends. The output is a demand forecast value for each product.

[0274] Step 3:

[0275] The server generates an order plan using an optimization algorithm based on predicted demand values. The input is the demand forecast, and the algorithm calculates the order quantity and purchase timing for each product, taking into account order lead time, inventory costs, and stockout risk. The output is the optimal order plan.

[0276] Step 4:

[0277] The server sends the generated order plan to the terminals in each store. Based on the received order plan, the terminals display appropriate sales promotion information on digital signage. For example, for products nearing their expiration date, the signage displays relevant recipes and discount information to encourage customer purchases.

[0278] Step 5:

[0279] Users provide actual sales results to the server. This feedback data, including sales quantity and time information, is used as input for analysis on the server. The data analysis evaluates the difference from the predicted results and manages it as output data to facilitate the retraining of the AI ​​model.

[0280] Step 6:

[0281] The server periodically retrains the generated AI model using feedback data. This involves using difference information as input data, updating the model, and improving its accuracy. The expected output is improved model accuracy in the next demand forecast.

[0282] (Application Example 1)

[0283] 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."

[0284] In modern distribution and retail industries, improving the accuracy of inventory management and demand forecasting is an urgent issue. The risk of waste due to excessive inventory and the loss of sales opportunities due to out-of-stock directly affect business efficiency and profits, so the need for a system that comprehensively solves these problems is increasing. Furthermore, a mechanism that enables store staff to efficiently obtain information and act is also required.

[0285] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0286] In this invention, the server includes means for acquiring sales history information, means for predicting demand based on the acquired sales history information, means for generating an optimal order plan based on the predicted demand, means for displaying the generated order plan, and means for providing the order plan and sales promotion information generated via a customer terminal to store users. As a result, efficient inventory management and accurate demand forecasting become possible, and optimization of store operations is realized.

[0287] The "sales history information" is information that records past sales data at each sales base, and is data that forms the basis for demand forecasting and inventory management.

[0288] The "means for predicting demand" is a technical element for analyzing the acquired sales history information and estimating future sales quantities and customer purchase trends.

[0289] The "order plan" indicates specific guidelines and plans for ordering necessary products in appropriate quantities within an appropriate period based on demand forecasting.

[0290] The "means for displaying" is a device or apparatus for outputting the generated information in a visually recognizable form, and plays a role in conveying important information to store staff and customers.

[0291] The "customer terminal" is an information device that can be operated by users who visit the store, and is used to receive necessary information and instructions.

[0292] "Sales promotion information" refers to information designed to encourage the purchase of a specific product, and mainly includes discounts, campaigns, and product benefits.

[0293] An "electronic display device" is a device that displays information in digital format and is used to visually present information within a store.

[0294] A "generative algorithm" is a series of computational procedures and methods used in information processing to analyze data and derive results.

[0295] A "prompt statement" is an input statement used to provide specific instructions or information to a generation algorithm in order to obtain the desired output.

[0296] As an embodiment for carrying out the present invention, the system has the following configuration. The server periodically acquires sales history information from each sales location. The sales history information is data on what products the store has sold and in what quantities in the past, and is basic data used for demand forecasting.

[0297] The server feeds the acquired sales history information into an AI model for demand forecasting. This AI model is built using machine learning libraries such as TensorFlow and PyTorch, and is continuously retrained to improve prediction accuracy. Based on the predicted demand, the server generates an optimal ordering plan. The ordering plan indicates which products to order, in what quantities, and at what time of year the orders should be placed.

[0298] The generated order plans and sales promotion information are visually displayed on electronic display devices within the store or provided to store users via customer terminals. This allows sales promotion information to be conveyed to customers in real time, making it possible to encourage the purchase of products that are at high risk of being discarded.

[0299] Furthermore, to evaluate the difference between actual sales data and predicted data and continuously improve the AI model, the server collects this difference information as feedback. For example, the server generates a prompt sentence "What are the products whose sales are likely to increase soon?" and realizes more accurate predictions by adjusting the algorithm.

[0300] With this system, users can easily obtain highly accurate demand predictions based on sales history information and the corresponding optimal ordering plans, thereby improving the efficiency of inventory management and sales promotion.

[0301] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0302] Step 1:

[0303] The server obtains sales history information from each sales base. The sales history information stores past sales data at the store in digital form. As input data, it includes product ID, sales date, sales quantity, etc. The obtained data is accumulated in the server's database.

[0304] Step 2:

[0305] The server uses the sales history information to make a demand prediction by the AI model. This model is constructed using TensorFlow and performs time series analysis. Sales history information is used as input, and predicted sales volume for the next certain period is generated as output. The server runs the AI model to obtain the prediction result.

[0306] Step 3:

[0307] The server generates an optimal ordering plan based on the predicted demand. The result of the demand prediction is used as input, and a plan including the ordering quantity and ordering timing for each product is created as output. This ordering plan also takes into account the inventory level and past ordering history.

[0308] Step 4:

[0309] Order plans and sales promotion information are transmitted from the server to electronic display devices and customer terminals in stores. Input data includes order plan information and sales promotion strategies, and output generates visual information to be provided to store staff and customers.

[0310] Step 5:

[0311] Users manage inventory and conduct sales promotion activities based on the displayed information. By operating the terminal and accessing sales promotion information, users can process inventory items and promote products nearing their expiration date.

[0312] Step 6:

[0313] The difference between actual sales data and forecast data is fed back to the server. This feedback allows the AI ​​model to be retrained, improving the accuracy of demand forecasting. The difference information is then used as input for further data processing, updating the AI ​​model to aim for more accurate forecasts.

[0314] Step 7:

[0315] The server continuously generates prompts to adjust the algorithm and tune the model. For example, it generates a prompt such as "Which products are likely to see increased sales next week?" to adjust the model and maximize operational effectiveness.

[0316] 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.

[0317] One possible embodiment of the present invention is the operation of a system incorporating an emotion engine. This system provides demand forecasting and optimized ordering plans based on sales history information, as well as a function to recognize user emotions and adjust sales promotion measures accordingly.

[0318] The server first retrieves sales history information from each store and uses an AI model to forecast demand based on that data. Based on the forecast results, the server generates an optimal ordering plan and sends this information to the store's terminal. For example, at store A, this month's sales history is sent to the server, and demand for the following month is forecasted.

[0319] The terminal displays received order plans to store staff and uses an emotion engine to recognize customers' emotions in real time. This emotion information is used, for example, for sales promotion through in-store digital signage. As a specific example, in store B, a camera scans the faces of customers who come in, and the emotion engine recognizes emotions such as joy or surprise from their facial expressions.

[0320] Based on the emotions it recognizes, the terminal adjusts sales promotion strategies. For example, it can provide discount information on products the customer has shown interest in, or suggest products that match a specific emotion. Store staff then use the suggestions from the terminal to optimize on-site customer service and product placement.

[0321] Furthermore, the server continuously collects actual sales results and forecast data, which is used to retrain the AI ​​model. The server is designed to improve the accuracy of demand forecasting and quantitatively improve the emotion engine algorithm. At store C, sales results and customer emotion reactions are sent to the server and used to improve forecast accuracy and adapt sales strategies for future events.

[0322] In this way, systems incorporating an emotion engine enable data-driven demand forecasting and real-time customer response, supporting store operations more effectively and efficiently.

[0323] The following describes the processing flow.

[0324] Step 1:

[0325] The server periodically collects sales history information from each store. The server integrates this data and stores it in a database. In this process, information such as the date of sale, product name, and quantity sold is organized.

[0326] Step 2:

[0327] The server uses an AI model to perform demand forecasting based on collected sales history information. The generated forecasts indicate the demand for each product over a certain period in the future, taking seasonality and trends into account.

[0328] Step 3:

[0329] The server generates an optimal ordering plan based on demand forecasts. The server considers product inventory levels and lead times to determine the necessary products and their order quantities. The ordering plan also includes adjustments to prevent excess inventory.

[0330] Step 4:

[0331] The server sends the order plan to terminals in each store. The terminals display this plan in a format that is easy for store staff to use, supporting quick confirmation and decision-making.

[0332] Step 5:

[0333] The device recognizes customers' emotions in real time through an emotion engine. The device acquires data from cameras and sensors installed in the store and analyzes emotions from the user's facial expressions and voice.

[0334] Step 6:

[0335] The terminal displays customized sales promotion strategies based on the customer's recognized emotional information. For example, if it determines that a customer is interested, it will display promotional content for related products on the digital signage.

[0336] Step 7:

[0337] Users (store staff) adjust sales strategies based on information provided by the terminal. Users utilize suggested discount information and sales promotion measures to optimize direct customer service and product placement.

[0338] Step 8:

[0339] The server collects the differences between actual sales data and the predictions used, and retrains the AI ​​model and emotion engine. The server analyzes this data and forms a feedback loop to improve the accuracy of the model in the future.

[0340] (Example 2)

[0341] 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".

[0342] Traditional sales systems focused on providing demand forecasts and order plans based on sales history information, but they lacked the ability to adjust sales promotion strategies based on individual customer sentiment, limiting their ability to optimize the customer experience. Furthermore, they lacked mechanisms for continuously improving the discrepancies between sales results and forecasts, which negatively impacted the accuracy of demand forecasts.

[0343] 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.

[0344] In this invention, the server includes means for collecting sales history data, means for forecasting demand based on the collected sales history data, and means for generating an optimized order plan based on the forecasted demand. This makes it possible to adjust sales promotion measures to take into account customer sentiment, thereby improving the accuracy of demand forecasting and enhancing the customer experience.

[0345] "Sales history data" refers to information such as transaction records and inventory fluctuations from past sales activities, and includes elements such as product category, sales quantity, and sales amount.

[0346] "Means of forecasting demand" refers to technical devices or methods for calculating and estimating future demand based on collected sales history data, and includes the use of machine learning models.

[0347] "Means for generating order plans" refers to technical devices or methods for creating plans to efficiently replenish goods based on predicted demand information.

[0348] A "device" refers to hardware used to visually provide generated information to a user, and includes display devices and mobile terminals.

[0349] "Means of recognizing customer emotions" refers to technologies that analyze a customer's facial expressions and actions to identify their emotional state, and includes combining cameras with emotion recognition software.

[0350] "Means of adjusting sales promotion strategies" refer to technical devices and methods that utilize customer sentiment information to formulate and execute the most appropriate sales activities and promotions at any given time.

[0351] An "artificial intelligence model" refers to a group of algorithms or programs that learn from large amounts of data and perform predictions and classifications, and includes the use of machine learning and deep learning technologies.

[0352] This invention relates to a system that integrates demand forecasting using sales history data with sales promotion that takes customer sentiment into consideration. This system, consisting of a server and terminals, streamlines sales activities and improves the customer experience.

[0353] The server is primarily responsible for data collection and analysis. First, the server collects sales history data from each store. This data includes sales volume, sales amount, and product category, and is automatically extracted from POS systems and inventory management systems. Using this data, the server predicts future demand using an AI model. Machine learning frameworks such as TensorFlow and PyTorch can be used for the AI ​​model. Based on the prediction results, the server generates an optimized ordering plan and sends it to the store's terminal.

[0354] The terminal visualizes received order plans for store staff. It also incorporates a camera and emotion recognition software, allowing it to scan customers' faces and recognize their emotions in real time. Emotion recognition software used includes tools like Amazon Rekognition, which identify emotions such as "joy" and "surprise" from customer facial expressions. Based on the recognized emotions, the terminal displays appropriate sales promotions on in-store displays and pop-ups. For example, if a customer shows a surprised expression, it can display discount information for that product.

[0355] Store staff, as users of the system, can use the information provided by the terminal to optimize customer service and product placement, thereby improving their actual sales activities. This enables data-driven decision-making, leading to more effective inventory management and increased customer satisfaction.

[0356] Furthermore, the server continuously monitors the discrepancies between sales results and forecast data, and uses this information to retrain the AI ​​model. This improves the accuracy of demand forecasting and enhances the effectiveness of promotions based on customer sentiment prediction data. Overall, the system aims to achieve both efficient sales management and high customer satisfaction.

[0357] As a concrete example, a prompt message for a generative AI model might be, "Based on customer sentiment data, please suggest the most effective discount strategy."

[0358] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0359] Step 1:

[0360] The server collects sales history data from each store. This data input includes information from POS systems and inventory management systems. The server analyzes this data and generates output that organizes and aggregates information such as sales volume, sales amount, and product category.

[0361] Step 2:

[0362] The server inputs collected sales history data into an AI model to perform demand forecasting. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to perform data calculations that estimate future demand based on patterns in past data. As a result, it outputs demand forecast data for the following month.

[0363] Step 3:

[0364] The server generates an optimized ordering plan based on demand forecast data. Here, demand data is used as input, and a supply chain management algorithm is applied to process the data to determine the quantity and timing of product replenishment. The output is a specific ordering schedule for each store.

[0365] Step 4:

[0366] The server sends the generated order plan to the store's terminal. The terminal receives the order plan from the server as input and displays it on the screen. This output includes a summary of the received order information and notification messages.

[0367] Step 5:

[0368] The terminal uses in-store cameras to scan customers' faces and recognize their emotions in real time. The input includes camera footage, which is then analyzed by emotion recognition APIs such as Amazon Rekognition. As a result, the recognized customer's emotion data is output.

[0369] Step 6:

[0370] The device adjusts sales promotion strategies based on recognized emotional data. Specifically, it takes customer emotions such as joy or surprise as input and displays appropriate discount information and promotional content on the screen. The output is the adjusted sales promotion information.

[0371] Step 7:

[0372] The server monitors the difference between actual sales data and forecast data. Using this difference data as input, the server performs data calculations to improve the accuracy of demand forecasting by retraining the AI ​​model. The output is the updated AI model parameters.

[0373] (Application Example 2)

[0374] 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."

[0375] Traditional store operations were limited to simple demand forecasting and ordering plans based on sales history data, and did not incorporate sales promotion strategies that considered consumers' real-time emotions. As a result, there were challenges in improving the customer experience and maximizing the effectiveness of sales promotions. Furthermore, the lack of continuous improvement of AI models based on sales results limited the accuracy of demand forecasting.

[0376] 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.

[0377] In this invention, the server includes means for acquiring sales history information, means for recognizing a consumer's facial expression and estimating their emotions, and means for generating sales promotion information based on the estimated emotions. This enables the implementation of sales promotion measures that reflect consumer emotions in real time, and continuous improvement of the AI ​​model based on sales results.

[0378] "Sales history information" refers to record data about past sales activities at each store, including information such as product name, quantity, date and time, and price.

[0379] "Demand forecasting" is a technique that uses statistical methods and AI models to estimate future sales volume and revenue based on past sales history information.

[0380] An "optimal ordering plan" is a strategy, including a list of products and quantities to be purchased, developed in response to predicted demand to minimize inventory costs and prevent stockouts.

[0381] "Methods for recognizing consumer facial expressions and estimating emotions" refers to technologies that use image processing techniques and machine learning algorithms to analyze emotions from facial images of consumers captured by a camera.

[0382] "Means of generating sales promotion information" refers to methods for determining and presenting marketing measures such as product discounts and purchase recommendations based on consumer sentiment and demand forecasts.

[0383] "Retraining an AI model" is the process of retraining an existing artificial intelligence model using actual sales data or newly acquired information to improve its prediction accuracy.

[0384] To implement this invention, the following system configuration is adopted. The server first acquires sales history information from each store and performs demand forecasting using an AI model. The main software used includes Python and TensorFlow. This analyzes the sales history data and predicts the number of products needed for the next period. Based on the forecast results, the server generates an optimal ordering plan and sends it to the terminals of each store.

[0385] The terminal not only displays received order plans to store staff, but also uses a camera to scan consumers' facial expressions in real time and recognize their emotions. Using a combination of OpenCV and TensorFlow, it classifies emotions into multiple categories such as joy and surprise, and generates sales promotion information based on this classification. This generated information is then presented to consumers via digital signage and smartphone displays.

[0386] Users (in this case, store staff) can adjust sales promotion strategies based on the generated emotional information. This enables personalized discount information and product recommendations for consumers. Furthermore, by feeding back newly acquired sales data and consumer emotional responses to the server, the AI ​​model is retrained, contributing to improved prediction accuracy.

[0387] For example, if a consumer smiles when looking at a particular product in a store, the terminal can display information such as "Buy this product now and get a 10% discount." Examples of prompts used for emotion recognition include "Present discount information when a smile is detected" and "Suggest related products when an expression of surprise is detected." In this way, appropriate responses based on consumer emotions become possible, contributing to improved store operational efficiency and increased customer satisfaction.

[0388] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0389] Step 1:

[0390] The server retrieves sales history information from each store. The input is sales history data for each individual store, and the output is integrated sales data. This data is collected using a Python script and stored in a database to prepare for future demand forecasting.

[0391] Step 2:

[0392] The server uses acquired sales history information to perform demand forecasting with an AI model. The input is integrated sales data, and the output is the predicted demand value. A TensorFlow model is used to analyze the data and generate estimated demand for each store.

[0393] Step 3:

[0394] The server generates an optimal ordering plan based on predicted demand. The input is predicted demand data, and the output is ordering plan data. An algorithm is used to calculate the order quantity and create the ordering plan.

[0395] Step 4:

[0396] The terminal displays the order plan received from the server to the store staff. The input is the order plan data, and the output is the information displayed on the screen. The terminal's GUI makes the information easy for staff to understand.

[0397] Step 5:

[0398] The device captures the customer's facial expressions using a camera in the store and recognizes their emotions. The input is the camera image, and the output is emotion data. Face detection is performed using OpenCV, and emotions are estimated using a TensorFlow model.

[0399] Step 6:

[0400] The device generates sales promotion information based on estimated emotions. The input is emotion data, and the output is sales promotion information. Based on the AI-generated promotional information, it provides consumers with product discounts and suggestions tailored to their needs.

[0401] Step 7:

[0402] Users adjust customer interactions based on the generated sales promotion information. The input is sales promotion strategy information, and the output is the content presented to the customer. Store staff make specific suggestions to customers, stimulating their purchasing intent.

[0403] Step 8:

[0404] The server collects actual sales data and consumer sentiment responses, and uses this data to retrain the AI ​​model. The input is sales results data and sentiment data, and the output is an improved AI model. The model is continuously updated to improve prediction accuracy.

[0405] 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.

[0406] 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.

[0407] 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.

[0408] [Third Embodiment]

[0409] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0410] 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.

[0411] 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).

[0412] 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.

[0413] 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.

[0414] 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).

[0415] 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.

[0416] 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.

[0417] 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.

[0418] 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.

[0419] 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.

[0420] 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".

[0421] The following system operation is conceivable as an embodiment of the present invention. This system predicts demand based on sales history information, enables the generation of order plans, and facilitates timely sales promotion.

[0422] The server periodically retrieves sales history information from each store and uses this data to perform demand forecasting with an AI model. For example, in store A, monthly sales history is sent to the server, where the data is analyzed by an AI algorithm to forecast demand for the following month.

[0423] The server generates an optimal ordering plan based on predicted demand. This ordering plan details purchase quantities and timings, and is optimized to minimize excess inventory and stockouts. The generated ordering plan is sent to each store's terminal, allowing store staff to use it to order their next products.

[0424] The terminals are connected to digital signage within the store and display promotional information, including products at high risk of being discarded. For example, a terminal in store B suggests recipes using ingredients nearing their expiration date and offers discounts on those products to appeal to customers. Users can then use this information to increase sales of their inventory.

[0425] The system works by having the server receive feedback on the difference between actual sales results and forecast data, and then periodically retraining the AI ​​model to improve its accuracy. At store C, information on the difference between forecasts and actual sales results is sent to the server and used to adjust the AI ​​model. In this way, the system is continuously improved, enabling more accurate demand forecasting.

[0426] This system enables store operators to make data-driven decisions, leading to more efficient inventory management and reduced food waste.

[0427] The following describes the processing flow.

[0428] Step 1:

[0429] The server retrieves sales history information from each store. The server works in conjunction with the stores' POS systems and automatically imports the most recent sales data into the database on a daily basis.

[0430] Step 2:

[0431] The server preprocesses the collected sales history information. The server identifies missing or outlier data, and then fills in and corrects them to generate a dataset suitable for the AI ​​model.

[0432] Step 3:

[0433] The server performs demand forecasting using an AI model. The server uses a pre-trained time-series forecasting model to predict future sales figures for each product.

[0434] Step 4:

[0435] The server generates an optimal ordering plan based on the demand forecast. Following the demand forecast, and taking inventory levels and lead times into consideration, the server calculates the quantity and timing of orders to be placed at the lowest possible cost.

[0436] Step 5:

[0437] The server sends the order plan to the terminals in each store. The terminals receive this order plan and prepare it for display in an easy-to-read format for the person in charge.

[0438] Step 6:

[0439] The terminal displays information about products at high risk of being discarded on digital signage. The terminal also displays recipes and discount information for identified products to appeal to customers.

[0440] Step 7:

[0441] Users (store staff) plan product orders and sales promotions based on information from the terminal. Users confirm the suggested quantities on the terminal, place actual orders, and implement promotional campaigns.

[0442] Step 8:

[0443] The server collects actual sales data and retrains the AI ​​model. The server compares the predicted data with the actual data and updates the model parameters to improve the accuracy of the AI ​​model.

[0444] (Example 1)

[0445] 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."

[0446] In today's distribution industry, proper inventory management and improved demand forecasting accuracy are crucial for solving problems such as excess inventory and stockouts. However, existing systems struggle to generate precise demand forecasts and optimal ordering plans that adapt to demand fluctuations. Furthermore, sales promotion measures are not adjusted in real time, resulting in inefficient sales of products with a high risk of spoilage. Solving these challenges is essential.

[0447] 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.

[0448] In this invention, the server includes means for collecting and pre-processing sales history information, means for predicting demand using an AI model generated based on the pre-processed sales history information, and means for generating an order plan using an optimization algorithm based on the predicted demand. This enables precise demand forecasting in response to demand fluctuations and the generation of an optimal order plan.

[0449] "Sales history information" refers to records of product sales at each store, including data such as product name, quantity sold, and date and time of sale.

[0450] A "generative AI model" is a system that uses machine learning algorithms to analyze data, identify patterns, and predict future demand.

[0451] "Preprocessing" refers to a series of processes, such as data organization, outlier removal, and missing value handling, that transform raw data into a format that can be easily handled by AI models.

[0452] "Demand forecasting" is the process of estimating future demand for a product based on past sales data and related information.

[0453] An "optimization algorithm" is a method for minimizing or maximizing resource utilization to achieve a specific objective, and in this context, it is used to optimize ordering plans in inventory management.

[0454] An "ordering plan" is a detailed plan that specifies the quantity and timing of purchases of goods in accordance with predicted demand, with the aim of minimizing the risk of excess inventory or stockouts.

[0455] "Sales promotion measures" are strategies and activities undertaken to increase product sales, aiming to communicate the appeal of a product to consumers and stimulate their desire to purchase.

[0456] "Feedback" refers to information obtained from actual sales results, and analyzing this data is crucial for improving the prediction accuracy of AI models.

[0457] This invention is a system that functions around three main elements: a server, a terminal, and a user. The server first collects sales history information from each store. This information includes details such as product name, quantity sold, and date and time of sale, and is stored in a database. The collected data is preprocessed and prepared in a format that is easy for the generated AI model to handle.

[0458] Next, the server uses a generative AI model to forecast demand. This AI model is designed based on machine learning algorithms and analyzes past sales data to estimate future demand. This provides a foundation for minimizing supply-demand mismatches. For example, it can make forecasts that take into account demand fluctuation patterns in line with specific seasons or events.

[0459] Furthermore, the server generates an order plan using an optimization algorithm based on the demand forecast results. This plan includes specific product quantities and order timings, and is designed to avoid excess inventory and stockouts while considering inventory costs. The order plan is sent to terminals, where store staff can use it to manage inventory efficiently.

[0460] The terminals are connected to digital signage used within the store, displaying sales promotion information in real time. This is particularly used to promote products at high risk of spoilage, aiming to increase sales by conveying the appeal of the products to customers. For example, for products nearing their expiration date, related recipes and discount information are displayed to stimulate purchasing intent.

[0461] Users provide actual sales results to the server, and this data is used as feedback to retrain the AI ​​model. This continuously improves the accuracy of the AI ​​model, enabling more accurate demand forecasting. Through this process, the system can automatically learn and evolve.

[0462] As a specific example, store D uses the prompt message, "Based on sales history over the past three months, predict the demand for a specific product for the next month," to perform demand forecasting. Based on this result, an ordering plan can be created, and the results can be used for promotions on terminals.

[0463] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0464] Step 1:

[0465] The server periodically collects sales history information from each store. This information is input as a dataset containing details such as product name, sales quantity, and sales date and time. The server processes this data by handling outliers and missing values, transforming it into well-formed data to generate preprocessed data suitable for input to the AI ​​model.

[0466] Step 2:

[0467] The server inputs pre-processed data into a generating AI model and performs demand forecasting. This process involves data calculations that recognize past sales patterns and use time series analysis to predict future consumption trends. The output is a demand forecast value for each product.

[0468] Step 3:

[0469] The server generates an order plan using an optimization algorithm based on predicted demand values. The input is the demand forecast, and the algorithm calculates the order quantity and purchase timing for each product, taking into account order lead time, inventory costs, and stockout risk. The output is the optimal order plan.

[0470] Step 4:

[0471] The server sends the generated order plan to the terminals in each store. Based on the received order plan, the terminals display appropriate sales promotion information on digital signage. For example, for products nearing their expiration date, the signage displays relevant recipes and discount information to encourage customer purchases.

[0472] Step 5:

[0473] Users provide actual sales results to the server. This feedback data, including sales quantity and time information, is used as input for analysis on the server. The data analysis evaluates the difference from the predicted results and manages it as output data to facilitate the retraining of the AI ​​model.

[0474] Step 6:

[0475] The server periodically retrains the generated AI model using feedback data. This involves using difference information as input data, updating the model, and improving its accuracy. The expected output is improved model accuracy in the next demand forecast.

[0476] (Application Example 1)

[0477] 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."

[0478] In today's distribution and retail industries, improving the accuracy of inventory management and demand forecasting is an urgent issue. The risk of waste due to excess inventory and lost sales opportunities due to stockouts directly impact operational efficiency and profits, thus increasing the need for systems that comprehensively address these issues. Furthermore, there is a demand for mechanisms that enable store staff to efficiently obtain information and take action.

[0479] 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.

[0480] In this invention, the server includes means for acquiring sales history information, means for predicting demand based on the acquired sales history information, means for generating an optimal order plan based on the predicted demand, means for displaying the generated order plan, and means for providing the generated order plan and sales promotion information to store users via customer terminals. This enables efficient inventory management and accurate demand forecasting, thereby optimizing store operations.

[0481] "Sales history information" refers to information that records past sales data at each sales location, and is the basis for demand forecasting and inventory management.

[0482] "Means of predicting demand" refer to technical elements that analyze acquired sales history information to estimate future sales volumes and customer purchasing trends.

[0483] An "ordering plan" is a set of specific guidelines and plans for ordering the necessary goods in the appropriate quantities within the appropriate timeframe, based on demand forecasts.

[0484] "Means of display" refers to equipment or devices that output generated information in a visually recognizable format, and plays a role in conveying important information to store staff and customers.

[0485] A "customer terminal" is an information device that can be operated by customers visiting a store, and is used to receive necessary information and instructions.

[0486] "Sales promotion information" refers to information designed to encourage the purchase of a specific product, and mainly includes discounts, campaigns, and product benefits.

[0487] An "electronic display device" is a device that displays information in digital format and is used to visually present information within a store.

[0488] A "generative algorithm" is a series of computational procedures and methods used in information processing to analyze data and derive results.

[0489] A "prompt statement" is an input statement used to provide specific instructions or information to a generation algorithm in order to obtain the desired output.

[0490] As an embodiment for carrying out the present invention, the system has the following configuration. The server periodically acquires sales history information from each sales location. The sales history information is data on what products the store has sold and in what quantities in the past, and is basic data used for demand forecasting.

[0491] The server feeds the acquired sales history information into an AI model for demand forecasting. This AI model is built using machine learning libraries such as TensorFlow and PyTorch, and is continuously retrained to improve prediction accuracy. Based on the predicted demand, the server generates an optimal ordering plan. The ordering plan indicates which products to order, in what quantities, and at what time of year the orders should be placed.

[0492] The generated order plans and sales promotion information are visually displayed on electronic display devices within the store or provided to store users via customer terminals. This allows sales promotion information to be conveyed to customers in real time, making it possible to encourage the purchase of products that are at high risk of being discarded.

[0493] Furthermore, the server collects this discrepancy information as feedback to evaluate the differences between actual sales data and forecast data, and to continuously improve the AI ​​model. For example, the server generates a prompt message such as "Which products are likely to see increased sales in the near future?" and adjusts the algorithm to achieve more accurate predictions.

[0494] This system allows users to easily obtain highly accurate demand forecasts based on sales history information and optimal ordering plans accordingly, thereby improving the efficiency of inventory management and sales promotion.

[0495] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0496] Step 1:

[0497] The server retrieves sales history information from each sales location. This sales history information is a digital record of past sales data at each store. Input data includes product ID, sales date, and sales quantity. The retrieved data is stored in the server's database.

[0498] Step 2:

[0499] The server uses sales history data to perform demand forecasting using an AI model. This model is built using TensorFlow and performs time series analysis. Sales history data is used as input, and the output generates predicted sales volume for the next specified period. The server runs the AI ​​model and obtains the forecast results.

[0500] Step 3:

[0501] The server generates an optimal ordering plan based on predicted demand. The input is the results of the demand forecast, and the output is a plan that includes the order quantity and timing for each product. This ordering plan also takes into account inventory levels and past ordering history.

[0502] Step 4:

[0503] Order plans and sales promotion information are transmitted from the server to electronic display devices and customer terminals in stores. Input data includes order plan information and sales promotion strategies, and output generates visual information to be provided to store staff and customers.

[0504] Step 5:

[0505] Users manage inventory and conduct sales promotion activities based on the displayed information. By operating the terminal and accessing sales promotion information, users can process inventory items and promote products nearing their expiration date.

[0506] Step 6:

[0507] The difference between actual sales data and forecast data is fed back to the server. This feedback allows the AI ​​model to be retrained, improving the accuracy of demand forecasting. The difference information is then used as input for further data processing, updating the AI ​​model to aim for more accurate forecasts.

[0508] Step 7:

[0509] The server continuously generates prompts to adjust the algorithm and tune the model. For example, it generates a prompt such as "Which products are likely to see increased sales next week?" to adjust the model and maximize operational effectiveness.

[0510] 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.

[0511] One possible embodiment of the present invention is the operation of a system incorporating an emotion engine. This system provides demand forecasting and optimized ordering plans based on sales history information, as well as a function to recognize user emotions and adjust sales promotion measures accordingly.

[0512] The server first retrieves sales history information from each store and uses an AI model to forecast demand based on that data. Based on the forecast results, the server generates an optimal ordering plan and sends this information to the store's terminal. For example, at store A, this month's sales history is sent to the server, and demand for the following month is forecasted.

[0513] The terminal displays received order plans to store staff and uses an emotion engine to recognize customers' emotions in real time. This emotion information is used, for example, for sales promotion through in-store digital signage. As a specific example, in store B, a camera scans the faces of customers who come in, and the emotion engine recognizes emotions such as joy or surprise from their facial expressions.

[0514] Based on the emotions it recognizes, the terminal adjusts sales promotion strategies. For example, it can provide discount information on products the customer has shown interest in, or suggest products that match a specific emotion. Store staff then use the suggestions from the terminal to optimize on-site customer service and product placement.

[0515] Furthermore, the server continuously collects actual sales results and forecast data, which is used to retrain the AI ​​model. The server is designed to improve the accuracy of demand forecasting and quantitatively improve the emotion engine algorithm. At store C, sales results and customer emotion reactions are sent to the server and used to improve forecast accuracy and adapt sales strategies for future events.

[0516] In this way, systems incorporating an emotion engine enable data-driven demand forecasting and real-time customer response, supporting store operations more effectively and efficiently.

[0517] The following describes the processing flow.

[0518] Step 1:

[0519] The server periodically collects sales history information from each store. The server integrates this data and stores it in a database. In this process, information such as the date of sale, product name, and quantity sold is organized.

[0520] Step 2:

[0521] The server uses an AI model to perform demand forecasting based on collected sales history information. The generated forecasts indicate the demand for each product over a certain period in the future, taking seasonality and trends into account.

[0522] Step 3:

[0523] The server generates an optimal ordering plan based on demand forecasts. The server considers product inventory levels and lead times to determine the necessary products and their order quantities. The ordering plan also includes adjustments to prevent excess inventory.

[0524] Step 4:

[0525] The server sends the order plan to terminals in each store. The terminals display this plan in a format that is easy for store staff to use, supporting quick confirmation and decision-making.

[0526] Step 5:

[0527] The device recognizes customers' emotions in real time through an emotion engine. The device acquires data from cameras and sensors installed in the store and analyzes emotions from the user's facial expressions and voice.

[0528] Step 6:

[0529] The terminal displays customized sales promotion strategies based on the customer's recognized emotional information. For example, if it determines that a customer is interested, it will display promotional content for related products on the digital signage.

[0530] Step 7:

[0531] Users (store staff) adjust sales strategies based on information provided by the terminal. Users utilize suggested discount information and sales promotion measures to optimize direct customer service and product placement.

[0532] Step 8:

[0533] The server collects the differences between actual sales data and the predictions used, and retrains the AI ​​model and emotion engine. The server analyzes this data and forms a feedback loop to improve the accuracy of the model in the future.

[0534] (Example 2)

[0535] 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."

[0536] Traditional sales systems focused on providing demand forecasts and order plans based on sales history information, but they lacked the ability to adjust sales promotion strategies based on individual customer sentiment, limiting their ability to optimize the customer experience. Furthermore, they lacked mechanisms for continuously improving the discrepancies between sales results and forecasts, which negatively impacted the accuracy of demand forecasts.

[0537] 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.

[0538] In this invention, the server includes means for collecting sales history data, means for forecasting demand based on the collected sales history data, and means for generating an optimized order plan based on the forecasted demand. This makes it possible to adjust sales promotion measures to take into account customer sentiment, thereby improving the accuracy of demand forecasting and enhancing the customer experience.

[0539] "Sales history data" refers to information such as transaction records and inventory fluctuations from past sales activities, and includes elements such as product category, sales quantity, and sales amount.

[0540] "Means of forecasting demand" refers to technical devices or methods for calculating and estimating future demand based on collected sales history data, and includes the use of machine learning models.

[0541] "Means for generating order plans" refers to technical devices or methods for creating plans to efficiently replenish goods based on predicted demand information.

[0542] A "device" refers to hardware used to visually provide generated information to a user, and includes display devices and mobile terminals.

[0543] "Means of recognizing customer emotions" refers to technologies that analyze a customer's facial expressions and actions to identify their emotional state, and includes combining cameras with emotion recognition software.

[0544] "Means of adjusting sales promotion strategies" refer to technical devices and methods that utilize customer sentiment information to formulate and execute the most appropriate sales activities and promotions at any given time.

[0545] An "artificial intelligence model" refers to a group of algorithms or programs that learn from large amounts of data and perform predictions and classifications, and includes the use of machine learning and deep learning technologies.

[0546] This invention relates to a system that integrates demand forecasting using sales history data with sales promotion that takes customer sentiment into consideration. This system, consisting of a server and terminals, streamlines sales activities and improves the customer experience.

[0547] The server is primarily responsible for data collection and analysis. First, the server collects sales history data from each store. This data includes sales volume, sales amount, and product category, and is automatically extracted from POS systems and inventory management systems. Using this data, the server predicts future demand using an AI model. Machine learning frameworks such as TensorFlow and PyTorch can be used for the AI ​​model. Based on the prediction results, the server generates an optimized ordering plan and sends it to the store's terminal.

[0548] The terminal visualizes received order plans for store staff. It also incorporates a camera and emotion recognition software, allowing it to scan customers' faces and recognize their emotions in real time. Emotion recognition software used includes tools like Amazon Rekognition, which identify emotions such as "joy" and "surprise" from customer facial expressions. Based on the recognized emotions, the terminal displays appropriate sales promotions on in-store displays and pop-ups. For example, if a customer shows a surprised expression, it can display discount information for that product.

[0549] Store staff, as users of the system, can use the information provided by the terminal to optimize customer service and product placement, thereby improving their actual sales activities. This enables data-driven decision-making, leading to more effective inventory management and increased customer satisfaction.

[0550] Furthermore, the server continuously monitors the discrepancies between sales results and forecast data, and uses this information to retrain the AI ​​model. This improves the accuracy of demand forecasting and enhances the effectiveness of promotions based on customer sentiment prediction data. Overall, the system aims to achieve both efficient sales management and high customer satisfaction.

[0551] As a concrete example, a prompt message for a generative AI model might be, "Based on customer sentiment data, please suggest the most effective discount strategy."

[0552] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0553] Step 1:

[0554] The server collects sales history data from each store. This data input includes information from POS systems and inventory management systems. The server analyzes this data and generates output that organizes and aggregates information such as sales volume, sales amount, and product category.

[0555] Step 2:

[0556] The server inputs collected sales history data into an AI model to perform demand forecasting. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to perform data calculations that estimate future demand based on patterns in past data. As a result, it outputs demand forecast data for the following month.

[0557] Step 3:

[0558] The server generates an optimized ordering plan based on demand forecast data. Here, demand data is used as input, and a supply chain management algorithm is applied to process the data to determine the quantity and timing of product replenishment. The output is a specific ordering schedule for each store.

[0559] Step 4:

[0560] The server sends the generated order plan to the store's terminal. The terminal receives the order plan from the server as input and displays it on the screen. This output includes a summary of the received order information and notification messages.

[0561] Step 5:

[0562] The terminal uses in-store cameras to scan customers' faces and recognize their emotions in real time. The input includes camera footage, which is then analyzed by emotion recognition APIs such as Amazon Rekognition. As a result, the recognized customer's emotion data is output.

[0563] Step 6:

[0564] The device adjusts sales promotion strategies based on recognized emotional data. Specifically, it takes customer emotions such as joy or surprise as input and displays appropriate discount information and promotional content on the screen. The output is the adjusted sales promotion information.

[0565] Step 7:

[0566] The server monitors the difference between actual sales data and forecast data. Using this difference data as input, the server performs data calculations to improve the accuracy of demand forecasting by retraining the AI ​​model. The output is the updated AI model parameters.

[0567] (Application Example 2)

[0568] 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."

[0569] Traditional store operations were limited to simple demand forecasting and ordering plans based on sales history data, and did not incorporate sales promotion strategies that considered consumers' real-time emotions. As a result, there were challenges in improving the customer experience and maximizing the effectiveness of sales promotions. Furthermore, the lack of continuous improvement of AI models based on sales results limited the accuracy of demand forecasting.

[0570] 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.

[0571] In this invention, the server includes means for acquiring sales history information, means for recognizing a consumer's facial expression and estimating their emotions, and means for generating sales promotion information based on the estimated emotions. This enables the implementation of sales promotion measures that reflect consumer emotions in real time, and continuous improvement of the AI ​​model based on sales results.

[0572] "Sales history information" refers to record data about past sales activities at each store, including information such as product name, quantity, date and time, and price.

[0573] "Demand forecasting" is a technique that uses statistical methods and AI models to estimate future sales volume and revenue based on past sales history information.

[0574] An "optimal ordering plan" is a strategy, including a list of products and quantities to be purchased, developed in response to predicted demand to minimize inventory costs and prevent stockouts.

[0575] "Methods for recognizing consumer facial expressions and estimating emotions" refers to technologies that use image processing techniques and machine learning algorithms to analyze emotions from facial images of consumers captured by a camera.

[0576] "Means of generating sales promotion information" refers to methods for determining and presenting marketing measures such as product discounts and purchase recommendations based on consumer sentiment and demand forecasts.

[0577] "Retraining an AI model" is the process of retraining an existing artificial intelligence model using actual sales data or newly acquired information to improve its prediction accuracy.

[0578] To implement this invention, the following system configuration is adopted. The server first acquires sales history information from each store and performs demand forecasting using an AI model. The main software used includes Python and TensorFlow. This analyzes the sales history data and predicts the number of products needed for the next period. Based on the forecast results, the server generates an optimal ordering plan and sends it to the terminals of each store.

[0579] The terminal not only displays received order plans to store staff, but also uses a camera to scan consumers' facial expressions in real time and recognize their emotions. Using a combination of OpenCV and TensorFlow, it classifies emotions into multiple categories such as joy and surprise, and generates sales promotion information based on this classification. This generated information is then presented to consumers via digital signage and smartphone displays.

[0580] Users (in this case, store staff) can adjust sales promotion strategies based on the generated emotional information. This enables personalized discount information and product recommendations for consumers. Furthermore, by feeding back newly acquired sales data and consumer emotional responses to the server, the AI ​​model is retrained, contributing to improved prediction accuracy.

[0581] For example, if a consumer smiles when looking at a particular product in a store, the terminal can display information such as "Buy this product now and get a 10% discount." Examples of prompts used for emotion recognition include "Present discount information when a smile is detected" and "Suggest related products when an expression of surprise is detected." In this way, appropriate responses based on consumer emotions become possible, contributing to improved store operational efficiency and increased customer satisfaction.

[0582] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0583] Step 1:

[0584] The server retrieves sales history information from each store. The input is sales history data for each individual store, and the output is integrated sales data. This data is collected using a Python script and stored in a database to prepare for future demand forecasting.

[0585] Step 2:

[0586] The server uses acquired sales history information to perform demand forecasting with an AI model. The input is integrated sales data, and the output is the predicted demand value. A TensorFlow model is used to analyze the data and generate estimated demand for each store.

[0587] Step 3:

[0588] The server generates an optimal ordering plan based on predicted demand. The input is predicted demand data, and the output is ordering plan data. An algorithm is used to calculate the order quantity and create the ordering plan.

[0589] Step 4:

[0590] The terminal displays the order plan received from the server to the store staff. The input is the order plan data, and the output is the information displayed on the screen. The terminal's GUI makes the information easy for staff to understand.

[0591] Step 5:

[0592] The device captures the customer's facial expressions using a camera in the store and recognizes their emotions. The input is the camera image, and the output is emotion data. Face detection is performed using OpenCV, and emotions are estimated using a TensorFlow model.

[0593] Step 6:

[0594] The device generates sales promotion information based on estimated emotions. The input is emotion data, and the output is sales promotion information. Based on the AI-generated promotional information, it provides consumers with product discounts and suggestions tailored to their needs.

[0595] Step 7:

[0596] Users adjust customer interactions based on the generated sales promotion information. The input is sales promotion strategy information, and the output is the content presented to the customer. Store staff make specific suggestions to customers, stimulating their purchasing intent.

[0597] Step 8:

[0598] The server collects actual sales data and consumer sentiment responses, and uses this data to retrain the AI ​​model. The input is sales results data and sentiment data, and the output is an improved AI model. The model is continuously updated to improve prediction accuracy.

[0599] 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.

[0600] 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.

[0601] 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.

[0602] [Fourth Embodiment]

[0603] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0604] 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.

[0605] 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).

[0606] 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.

[0607] 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.

[0608] 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).

[0609] 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.

[0610] 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.

[0611] 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.

[0612] 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.

[0613] 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.

[0614] 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.

[0615] 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".

[0616] The following system operation is conceivable as an embodiment of the present invention. This system predicts demand based on sales history information, enables the generation of order plans, and facilitates timely sales promotion.

[0617] The server periodically retrieves sales history information from each store and uses this data to perform demand forecasting with an AI model. For example, in store A, monthly sales history is sent to the server, where the data is analyzed by an AI algorithm to forecast demand for the following month.

[0618] The server generates an optimal ordering plan based on predicted demand. This ordering plan details purchase quantities and timings, and is optimized to minimize excess inventory and stockouts. The generated ordering plan is sent to each store's terminal, allowing store staff to use it to order their next products.

[0619] The terminals are connected to digital signage within the store and display promotional information, including products at high risk of being discarded. For example, a terminal in store B suggests recipes using ingredients nearing their expiration date and offers discounts on those products to appeal to customers. Users can then use this information to increase sales of their inventory.

[0620] The system works by having the server receive feedback on the difference between actual sales results and forecast data, and then periodically retraining the AI ​​model to improve its accuracy. At store C, information on the difference between forecasts and actual sales results is sent to the server and used to adjust the AI ​​model. In this way, the system is continuously improved, enabling more accurate demand forecasting.

[0621] This system enables store operators to make data-driven decisions, leading to more efficient inventory management and reduced food waste.

[0622] The following describes the processing flow.

[0623] Step 1:

[0624] The server retrieves sales history information from each store. The server works in conjunction with the stores' POS systems and automatically imports the most recent sales data into the database on a daily basis.

[0625] Step 2:

[0626] The server preprocesses the collected sales history information. The server identifies missing or outlier data, and then fills in and corrects them to generate a dataset suitable for the AI ​​model.

[0627] Step 3:

[0628] The server performs demand forecasting using an AI model. The server uses a pre-trained time-series forecasting model to predict future sales figures for each product.

[0629] Step 4:

[0630] The server generates an optimal ordering plan based on the demand forecast. Following the demand forecast, and taking inventory levels and lead times into consideration, the server calculates the quantity and timing of orders to be placed at the lowest possible cost.

[0631] Step 5:

[0632] The server sends the order plan to the terminals in each store. The terminals receive this order plan and prepare it for display in an easy-to-read format for the person in charge.

[0633] Step 6:

[0634] The terminal displays information about products at high risk of being discarded on digital signage. The terminal also displays recipes and discount information for identified products to appeal to customers.

[0635] Step 7:

[0636] Users (store staff) plan product orders and sales promotions based on information from the terminal. Users confirm the suggested quantities on the terminal, place actual orders, and implement promotional campaigns.

[0637] Step 8:

[0638] The server collects actual sales data and retrains the AI ​​model. The server compares the predicted data with the actual data and updates the model parameters to improve the accuracy of the AI ​​model.

[0639] (Example 1)

[0640] 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".

[0641] In today's distribution industry, proper inventory management and improved demand forecasting accuracy are crucial for solving problems such as excess inventory and stockouts. However, existing systems struggle to generate precise demand forecasts and optimal ordering plans that adapt to demand fluctuations. Furthermore, sales promotion measures are not adjusted in real time, resulting in inefficient sales of products with a high risk of spoilage. Solving these challenges is essential.

[0642] 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.

[0643] In this invention, the server includes means for collecting and pre-processing sales history information, means for predicting demand using an AI model generated based on the pre-processed sales history information, and means for generating an order plan using an optimization algorithm based on the predicted demand. This enables precise demand forecasting in response to demand fluctuations and the generation of an optimal order plan.

[0644] "Sales history information" refers to records of product sales at each store, including data such as product name, quantity sold, and date and time of sale.

[0645] A "generative AI model" is a system that uses machine learning algorithms to analyze data, identify patterns, and predict future demand.

[0646] "Preprocessing" refers to a series of processes, such as data organization, outlier removal, and missing value handling, that transform raw data into a format that can be easily handled by AI models.

[0647] "Demand forecasting" is the process of estimating future demand for a product based on past sales data and related information.

[0648] An "optimization algorithm" is a method for minimizing or maximizing resource utilization to achieve a specific objective, and in this context, it is used to optimize ordering plans in inventory management.

[0649] An "ordering plan" is a detailed plan that specifies the quantity and timing of purchases of goods in accordance with predicted demand, with the aim of minimizing the risk of excess inventory or stockouts.

[0650] "Sales promotion measures" are strategies and activities undertaken to increase product sales, aiming to communicate the appeal of a product to consumers and stimulate their desire to purchase.

[0651] "Feedback" refers to information obtained from actual sales results, and analyzing this data is crucial for improving the prediction accuracy of AI models.

[0652] This invention is a system that functions around three main elements: a server, a terminal, and a user. The server first collects sales history information from each store. This information includes details such as product name, quantity sold, and date and time of sale, and is stored in a database. The collected data is preprocessed and prepared in a format that is easy for the generated AI model to handle.

[0653] Next, the server uses a generative AI model to forecast demand. This AI model is designed based on machine learning algorithms and analyzes past sales data to estimate future demand. This provides a foundation for minimizing supply-demand mismatches. For example, it can make forecasts that take into account demand fluctuation patterns in line with specific seasons or events.

[0654] Furthermore, the server generates an order plan using an optimization algorithm based on the demand forecast results. This plan includes specific product quantities and order timings, and is designed to avoid excess inventory and stockouts while considering inventory costs. The order plan is sent to terminals, where store staff can use it to manage inventory efficiently.

[0655] The terminals are connected to digital signage used within the store, displaying sales promotion information in real time. This is particularly used to promote products at high risk of spoilage, aiming to increase sales by conveying the appeal of the products to customers. For example, for products nearing their expiration date, related recipes and discount information are displayed to stimulate purchasing intent.

[0656] Users provide actual sales results to the server, and this data is used as feedback to retrain the AI ​​model. This continuously improves the accuracy of the AI ​​model, enabling more accurate demand forecasting. Through this process, the system can automatically learn and evolve.

[0657] As a specific example, store D uses the prompt message, "Based on sales history over the past three months, predict the demand for a specific product for the next month," to perform demand forecasting. Based on this result, an ordering plan can be created, and the results can be used for promotions on terminals.

[0658] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0659] Step 1:

[0660] The server periodically collects sales history information from each store. This information is input as a dataset containing details such as product name, sales quantity, and sales date and time. The server processes this data by handling outliers and missing values, transforming it into well-formed data to generate preprocessed data suitable for input to the AI ​​model.

[0661] Step 2:

[0662] The server inputs pre-processed data into a generating AI model and performs demand forecasting. This process involves data calculations that recognize past sales patterns and use time series analysis to predict future consumption trends. The output is a demand forecast value for each product.

[0663] Step 3:

[0664] The server generates an order plan using an optimization algorithm based on predicted demand values. The input is the demand forecast, and the algorithm calculates the order quantity and purchase timing for each product, taking into account order lead time, inventory costs, and stockout risk. The output is the optimal order plan.

[0665] Step 4:

[0666] The server sends the generated order plan to the terminals in each store. Based on the received order plan, the terminals display appropriate sales promotion information on digital signage. For example, for products nearing their expiration date, the signage displays relevant recipes and discount information to encourage customer purchases.

[0667] Step 5:

[0668] Users provide actual sales results to the server. This feedback data, including sales quantity and time information, is used as input for analysis on the server. The data analysis evaluates the difference from the predicted results and manages it as output data to facilitate the retraining of the AI ​​model.

[0669] Step 6:

[0670] The server periodically retrains the generated AI model using feedback data. This involves using difference information as input data, updating the model, and improving its accuracy. The expected output is improved model accuracy in the next demand forecast.

[0671] (Application Example 1)

[0672] 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".

[0673] In today's distribution and retail industries, improving the accuracy of inventory management and demand forecasting is an urgent issue. The risk of waste due to excess inventory and lost sales opportunities due to stockouts directly impact operational efficiency and profits, thus increasing the need for systems that comprehensively address these issues. Furthermore, there is a demand for mechanisms that enable store staff to efficiently obtain information and take action.

[0674] 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.

[0675] In this invention, the server includes means for acquiring sales history information, means for predicting demand based on the acquired sales history information, means for generating an optimal order plan based on the predicted demand, means for displaying the generated order plan, and means for providing the generated order plan and sales promotion information to store users via customer terminals. This enables efficient inventory management and accurate demand forecasting, thereby optimizing store operations.

[0676] "Sales history information" refers to information that records past sales data at each sales location, and is the basis for demand forecasting and inventory management.

[0677] "Means of predicting demand" refer to technical elements that analyze acquired sales history information to estimate future sales volumes and customer purchasing trends.

[0678] An "ordering plan" is a set of specific guidelines and plans for ordering the necessary goods in the appropriate quantities within the appropriate timeframe, based on demand forecasts.

[0679] "Means of display" refers to equipment or devices that output generated information in a visually recognizable format, and plays a role in conveying important information to store staff and customers.

[0680] A "customer terminal" is an information device that can be operated by customers visiting a store, and is used to receive necessary information and instructions.

[0681] "Sales promotion information" refers to information designed to encourage the purchase of a specific product, and mainly includes discounts, campaigns, and product benefits.

[0682] An "electronic display device" is a device that displays information in digital format and is used to visually present information within a store.

[0683] A "generative algorithm" is a series of computational procedures and methods used in information processing to analyze data and derive results.

[0684] A "prompt statement" is an input statement used to provide specific instructions or information to a generation algorithm in order to obtain the desired output.

[0685] As an embodiment for carrying out the present invention, the system has the following configuration. The server periodically acquires sales history information from each sales location. The sales history information is data on what products the store has sold and in what quantities in the past, and is basic data used for demand forecasting.

[0686] The server feeds the acquired sales history information into an AI model for demand forecasting. This AI model is built using machine learning libraries such as TensorFlow and PyTorch, and is continuously retrained to improve prediction accuracy. Based on the predicted demand, the server generates an optimal ordering plan. The ordering plan indicates which products to order, in what quantities, and at what time of year the orders should be placed.

[0687] The generated order plans and sales promotion information are visually displayed on electronic display devices within the store or provided to store users via customer terminals. This allows sales promotion information to be conveyed to customers in real time, making it possible to encourage the purchase of products that are at high risk of being discarded.

[0688] Furthermore, the server collects this discrepancy information as feedback to evaluate the differences between actual sales data and forecast data, and to continuously improve the AI ​​model. For example, the server generates a prompt message such as "Which products are likely to see increased sales in the near future?" and adjusts the algorithm to achieve more accurate predictions.

[0689] This system allows users to easily obtain highly accurate demand forecasts based on sales history information and optimal ordering plans accordingly, thereby improving the efficiency of inventory management and sales promotion.

[0690] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0691] Step 1:

[0692] The server retrieves sales history information from each sales location. This sales history information is a digital record of past sales data at each store. Input data includes product ID, sales date, and sales quantity. The retrieved data is stored in the server's database.

[0693] Step 2:

[0694] The server uses sales history data to perform demand forecasting using an AI model. This model is built using TensorFlow and performs time series analysis. Sales history data is used as input, and the output generates predicted sales volume for the next specified period. The server runs the AI ​​model and obtains the forecast results.

[0695] Step 3:

[0696] The server generates an optimal ordering plan based on predicted demand. The input is the results of the demand forecast, and the output is a plan that includes the order quantity and timing for each product. This ordering plan also takes into account inventory levels and past ordering history.

[0697] Step 4:

[0698] Order plans and sales promotion information are transmitted from the server to electronic display devices and customer terminals in stores. Input data includes order plan information and sales promotion strategies, and output generates visual information to be provided to store staff and customers.

[0699] Step 5:

[0700] Users manage inventory and conduct sales promotion activities based on the displayed information. By operating the terminal and accessing sales promotion information, users can process inventory items and promote products nearing their expiration date.

[0701] Step 6:

[0702] The difference between actual sales data and forecast data is fed back to the server. This feedback allows the AI ​​model to be retrained, improving the accuracy of demand forecasting. The difference information is then used as input for further data processing, updating the AI ​​model to aim for more accurate forecasts.

[0703] Step 7:

[0704] The server continuously generates prompts to adjust the algorithm and tune the model. For example, it generates a prompt such as "Which products are likely to see increased sales next week?" to adjust the model and maximize operational effectiveness.

[0705] 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.

[0706] One possible embodiment of the present invention is the operation of a system incorporating an emotion engine. This system provides demand forecasting and optimized ordering plans based on sales history information, as well as a function to recognize user emotions and adjust sales promotion measures accordingly.

[0707] The server first retrieves sales history information from each store and uses an AI model to forecast demand based on that data. Based on the forecast results, the server generates an optimal ordering plan and sends this information to the store's terminal. For example, at store A, this month's sales history is sent to the server, and demand for the following month is forecasted.

[0708] The terminal displays received order plans to store staff and uses an emotion engine to recognize customers' emotions in real time. This emotion information is used, for example, for sales promotion through in-store digital signage. As a specific example, in store B, a camera scans the faces of customers who come in, and the emotion engine recognizes emotions such as joy or surprise from their facial expressions.

[0709] Based on the emotions it recognizes, the terminal adjusts sales promotion strategies. For example, it can provide discount information on products the customer has shown interest in, or suggest products that match a specific emotion. Store staff then use the suggestions from the terminal to optimize on-site customer service and product placement.

[0710] Furthermore, the server continuously collects actual sales results and forecast data, which is used to retrain the AI ​​model. The server is designed to improve the accuracy of demand forecasting and quantitatively improve the emotion engine algorithm. At store C, sales results and customer emotion reactions are sent to the server and used to improve forecast accuracy and adapt sales strategies for future events.

[0711] In this way, systems incorporating an emotion engine enable data-driven demand forecasting and real-time customer response, supporting store operations more effectively and efficiently.

[0712] The following describes the processing flow.

[0713] Step 1:

[0714] The server periodically collects sales history information from each store. The server integrates this data and stores it in a database. In this process, information such as the date of sale, product name, and quantity sold is organized.

[0715] Step 2:

[0716] The server uses an AI model to perform demand forecasting based on collected sales history information. The generated forecasts indicate the demand for each product over a certain period in the future, taking seasonality and trends into account.

[0717] Step 3:

[0718] The server generates an optimal ordering plan based on demand forecasts. The server considers product inventory levels and lead times to determine the necessary products and their order quantities. The ordering plan also includes adjustments to prevent excess inventory.

[0719] Step 4:

[0720] The server sends the order plan to terminals in each store. The terminals display this plan in a format that is easy for store staff to use, supporting quick confirmation and decision-making.

[0721] Step 5:

[0722] The device recognizes customers' emotions in real time through an emotion engine. The device acquires data from cameras and sensors installed in the store and analyzes emotions from the user's facial expressions and voice.

[0723] Step 6:

[0724] The terminal displays customized sales promotion strategies based on the customer's recognized emotional information. For example, if it determines that a customer is interested, it will display promotional content for related products on the digital signage.

[0725] Step 7:

[0726] Users (store staff) adjust sales strategies based on information provided by the terminal. Users utilize suggested discount information and sales promotion measures to optimize direct customer service and product placement.

[0727] Step 8:

[0728] The server collects the differences between actual sales data and the predictions used, and retrains the AI ​​model and emotion engine. The server analyzes this data and forms a feedback loop to improve the accuracy of the model in the future.

[0729] (Example 2)

[0730] 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".

[0731] Traditional sales systems focused on providing demand forecasts and order plans based on sales history information, but they lacked the ability to adjust sales promotion strategies based on individual customer sentiment, limiting their ability to optimize the customer experience. Furthermore, they lacked mechanisms for continuously improving the discrepancies between sales results and forecasts, which negatively impacted the accuracy of demand forecasts.

[0732] 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.

[0733] In this invention, the server includes means for collecting sales history data, means for forecasting demand based on the collected sales history data, and means for generating an optimized order plan based on the forecasted demand. This makes it possible to adjust sales promotion measures to take into account customer sentiment, thereby improving the accuracy of demand forecasting and enhancing the customer experience.

[0734] "Sales history data" refers to information such as transaction records and inventory fluctuations from past sales activities, and includes elements such as product category, sales quantity, and sales amount.

[0735] "Means of forecasting demand" refers to technical devices or methods for calculating and estimating future demand based on collected sales history data, and includes the use of machine learning models.

[0736] "Means for generating order plans" refers to technical devices or methods for creating plans to efficiently replenish goods based on predicted demand information.

[0737] A "device" refers to hardware used to visually provide generated information to a user, and includes display devices and mobile terminals.

[0738] "Means of recognizing customer emotions" refers to technologies that analyze a customer's facial expressions and actions to identify their emotional state, and includes combining cameras with emotion recognition software.

[0739] "Means of adjusting sales promotion strategies" refer to technical devices and methods that utilize customer sentiment information to formulate and execute the most appropriate sales activities and promotions at any given time.

[0740] An "artificial intelligence model" refers to a group of algorithms or programs that learn from large amounts of data and perform predictions and classifications, and includes the use of machine learning and deep learning technologies.

[0741] This invention relates to a system that integrates demand forecasting using sales history data with sales promotion that takes customer sentiment into consideration. This system, consisting of a server and terminals, streamlines sales activities and improves the customer experience.

[0742] The server is primarily responsible for data collection and analysis. First, the server collects sales history data from each store. This data includes sales volume, sales amount, and product category, and is automatically extracted from POS systems and inventory management systems. Using this data, the server predicts future demand using an AI model. Machine learning frameworks such as TensorFlow and PyTorch can be used for the AI ​​model. Based on the prediction results, the server generates an optimized ordering plan and sends it to the store's terminal.

[0743] The terminal visualizes received order plans for store staff. It also incorporates a camera and emotion recognition software, allowing it to scan customers' faces and recognize their emotions in real time. Emotion recognition software used includes tools like Amazon Rekognition, which identify emotions such as "joy" and "surprise" from customer facial expressions. Based on the recognized emotions, the terminal displays appropriate sales promotions on in-store displays and pop-ups. For example, if a customer shows a surprised expression, it can display discount information for that product.

[0744] Store staff, as users of the system, can use the information provided by the terminal to optimize customer service and product placement, thereby improving their actual sales activities. This enables data-driven decision-making, leading to more effective inventory management and increased customer satisfaction.

[0745] Furthermore, the server continuously monitors the discrepancies between sales results and forecast data, and uses this information to retrain the AI ​​model. This improves the accuracy of demand forecasting and enhances the effectiveness of promotions based on customer sentiment prediction data. Overall, the system aims to achieve both efficient sales management and high customer satisfaction.

[0746] As a concrete example, a prompt message for a generative AI model might be, "Based on customer sentiment data, please suggest the most effective discount strategy."

[0747] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0748] Step 1:

[0749] The server collects sales history data from each store. This data input includes information from POS systems and inventory management systems. The server analyzes this data and generates output that organizes and aggregates information such as sales volume, sales amount, and product category.

[0750] Step 2:

[0751] The server inputs collected sales history data into an AI model to perform demand forecasting. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to perform data calculations that estimate future demand based on patterns in past data. As a result, it outputs demand forecast data for the following month.

[0752] Step 3:

[0753] The server generates an optimized ordering plan based on demand forecast data. Here, demand data is used as input, and a supply chain management algorithm is applied to process the data to determine the quantity and timing of product replenishment. The output is a specific ordering schedule for each store.

[0754] Step 4:

[0755] The server sends the generated order plan to the store's terminal. The terminal receives the order plan from the server as input and displays it on the screen. This output includes a summary of the received order information and notification messages.

[0756] Step 5:

[0757] The terminal uses in-store cameras to scan customers' faces and recognize their emotions in real time. The input includes camera footage, which is then analyzed by emotion recognition APIs such as Amazon Rekognition. As a result, the recognized customer's emotion data is output.

[0758] Step 6:

[0759] The device adjusts sales promotion strategies based on recognized emotional data. Specifically, it takes customer emotions such as joy or surprise as input and displays appropriate discount information and promotional content on the screen. The output is the adjusted sales promotion information.

[0760] Step 7:

[0761] The server monitors the difference between actual sales data and forecast data. Using this difference data as input, the server performs data calculations to improve the accuracy of demand forecasting by retraining the AI ​​model. The output is the updated AI model parameters.

[0762] (Application Example 2)

[0763] 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".

[0764] Traditional store operations were limited to simple demand forecasting and ordering plans based on sales history data, and did not incorporate sales promotion strategies that considered consumers' real-time emotions. As a result, there were challenges in improving the customer experience and maximizing the effectiveness of sales promotions. Furthermore, the lack of continuous improvement of AI models based on sales results limited the accuracy of demand forecasting.

[0765] 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.

[0766] In this invention, the server includes means for acquiring sales history information, means for recognizing a consumer's facial expression and estimating their emotions, and means for generating sales promotion information based on the estimated emotions. This enables the implementation of sales promotion measures that reflect consumer emotions in real time, and continuous improvement of the AI ​​model based on sales results.

[0767] "Sales history information" refers to record data about past sales activities at each store, including information such as product name, quantity, date and time, and price.

[0768] "Demand forecasting" is a technique that uses statistical methods and AI models to estimate future sales volume and revenue based on past sales history information.

[0769] An "optimal ordering plan" is a strategy, including a list of products and quantities to be purchased, developed in response to predicted demand to minimize inventory costs and prevent stockouts.

[0770] "Methods for recognizing consumer facial expressions and estimating emotions" refers to technologies that use image processing techniques and machine learning algorithms to analyze emotions from facial images of consumers captured by a camera.

[0771] "Means of generating sales promotion information" refers to methods for determining and presenting marketing measures such as product discounts and purchase recommendations based on consumer sentiment and demand forecasts.

[0772] "Retraining an AI model" is the process of retraining an existing artificial intelligence model using actual sales data or newly acquired information to improve its prediction accuracy.

[0773] To implement this invention, the following system configuration is adopted. The server first acquires sales history information from each store and performs demand forecasting using an AI model. The main software used includes Python and TensorFlow. This analyzes the sales history data and predicts the number of products needed for the next period. Based on the forecast results, the server generates an optimal ordering plan and sends it to the terminals of each store.

[0774] The terminal not only displays received order plans to store staff, but also uses a camera to scan consumers' facial expressions in real time and recognize their emotions. Using a combination of OpenCV and TensorFlow, it classifies emotions into multiple categories such as joy and surprise, and generates sales promotion information based on this classification. This generated information is then presented to consumers via digital signage and smartphone displays.

[0775] Users (in this case, store staff) can adjust sales promotion strategies based on the generated emotional information. This enables personalized discount information and product recommendations for consumers. Furthermore, by feeding back newly acquired sales data and consumer emotional responses to the server, the AI ​​model is retrained, contributing to improved prediction accuracy.

[0776] For example, if a consumer smiles when looking at a particular product in a store, the terminal can display information such as "Buy this product now and get a 10% discount." Examples of prompts used for emotion recognition include "Present discount information when a smile is detected" and "Suggest related products when an expression of surprise is detected." In this way, appropriate responses based on consumer emotions become possible, contributing to improved store operational efficiency and increased customer satisfaction.

[0777] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0778] Step 1:

[0779] The server retrieves sales history information from each store. The input is sales history data for each individual store, and the output is integrated sales data. This data is collected using a Python script and stored in a database to prepare for future demand forecasting.

[0780] Step 2:

[0781] The server uses acquired sales history information to perform demand forecasting with an AI model. The input is integrated sales data, and the output is the predicted demand value. A TensorFlow model is used to analyze the data and generate estimated demand for each store.

[0782] Step 3:

[0783] The server generates an optimal ordering plan based on predicted demand. The input is predicted demand data, and the output is ordering plan data. An algorithm is used to calculate the order quantity and create the ordering plan.

[0784] Step 4:

[0785] The terminal displays the order plan received from the server to the store staff. The input is the order plan data, and the output is the information displayed on the screen. The terminal's GUI makes the information easy for staff to understand.

[0786] Step 5:

[0787] The device captures the customer's facial expressions using a camera in the store and recognizes their emotions. The input is the camera image, and the output is emotion data. Face detection is performed using OpenCV, and emotions are estimated using a TensorFlow model.

[0788] Step 6:

[0789] The device generates sales promotion information based on estimated emotions. The input is emotion data, and the output is sales promotion information. Based on the AI-generated promotional information, it provides consumers with product discounts and suggestions tailored to their needs.

[0790] Step 7:

[0791] Users adjust customer interactions based on the generated sales promotion information. The input is sales promotion strategy information, and the output is the content presented to the customer. Store staff make specific suggestions to customers, stimulating their purchasing intent.

[0792] Step 8:

[0793] The server collects actual sales data and consumer sentiment responses, and uses this data to retrain the AI ​​model. The input is sales results data and sentiment data, and the output is an improved AI model. The model is continuously updated to improve prediction accuracy.

[0794] 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.

[0795] 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.

[0796] 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 robot 414.

[0797] 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.

[0798] 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.

[0799] 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.

[0800] 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.

[0801] 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.

[0802] 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."

[0803] 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.

[0804] 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.

[0805] 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.

[0806] 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.

[0807] 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.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] 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.

[0812] 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.

[0813] 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.

[0814] 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.

[0815] The following is further disclosed regarding the embodiments described above.

[0816] (Claim 1)

[0817] Means of obtaining sales history information,

[0818] A means of predicting demand based on acquired sales history information,

[0819] A means for generating an optimal ordering plan based on predicted demand,

[0820] A means for displaying the generated order plan,

[0821] A system that includes this.

[0822] (Claim 2)

[0823] A means of identifying products at high risk of being discarded based on demand forecasts,

[0824] A means of proposing sales promotion measures for identified products,

[0825] The system according to claim 1, including the following:

[0826] (Claim 3)

[0827] Evaluate the difference between actual sales data and forecast data.

[0828] Methods for retraining AI models to improve the accuracy of demand forecasting,

[0829] The system according to claim 1, including the following:

[0830] "Example 1"

[0831] (Claim 1)

[0832] A means for collecting and pre-processing sales history information,

[0833] A means for predicting demand using a generated AI model based on pre-processed sales history information,

[0834] A means for generating an order plan using an optimization algorithm based on predicted demand,

[0835] A means of displaying and distributing the generated order plan,

[0836] A method for evaluating actual sales data and retraining the AI ​​model,

[0837] A system that includes this.

[0838] (Claim 2)

[0839] A means of identifying products with a high risk of being discarded based on demand forecasts, and proposing sales promotion measures for those identified products,

[0840] A means of displaying sales promotion information using a terminal,

[0841] The system according to claim 1, including the following:

[0842] (Claim 3)

[0843] A method for periodically retraining the AI ​​model using feedback to improve the accuracy of the generated AI model by analyzing the differences between actual sales data and forecast data,

[0844] The system according to claim 1, including the following:

[0845] "Application Example 1"

[0846] (Claim 1)

[0847] Means of obtaining sales history information,

[0848] A means of predicting demand based on acquired sales history information,

[0849] A means for generating an optimal ordering plan based on predicted demand,

[0850] A means for displaying the generated order plan,

[0851] A means of providing store users with order plans and sales promotion information generated via customer terminals,

[0852] A system that includes this.

[0853] (Claim 2)

[0854] A means of identifying products at high risk of being discarded based on demand forecasts,

[0855] Proposing sales promotion measures for identified products and a means of displaying them on an electronic display device,

[0856] The system according to claim 1, including the following:

[0857] (Claim 3)

[0858] Evaluate the difference between actual sales data and forecast data.

[0859] A means of retraining generative algorithms to improve the accuracy of demand forecasting,

[0860] A means for generating command statements for a demand forecasting model based on instructions from an external input and adjusting the model,

[0861] The system according to claim 1, including the following:

[0862] "Example 2 of combining an emotion engine"

[0863] (Claim 1)

[0864] Means for collecting sales history data,

[0865] A means of predicting demand based on collected sales history data,

[0866] Means for generating an optimized ordering plan based on predicted demand,

[0867] A device that displays the generated order plan,

[0868] Means of recognizing customer emotions,

[0869] A means of adjusting sales promotion measures based on perceived emotions,

[0870] A system that includes this.

[0871] (Claim 2)

[0872] The system according to claim 1, which proposes sales promotion measures for a specific product based on customer sentiment information.

[0873] (Claim 3)

[0874] Evaluate the difference between actual sales data and forecast data.

[0875] A means of retraining artificial intelligence models to improve the accuracy of demand forecasting,

[0876] The system according to claim 1, including the following:

[0877] "Application example 2 when combining with an emotional engine"

[0878] (Claim 1)

[0879] Means of obtaining sales history information,

[0880] A means of predicting demand based on acquired sales history information,

[0881] A means for generating an optimal ordering plan based on predicted demand,

[0882] A means for displaying the generated order plan,

[0883] A means of recognizing consumers' facial expressions and estimating their emotions,

[0884] A means for generating sales promotion information based on estimated emotions,

[0885] A system that includes this.

[0886] (Claim 2)

[0887] A means of identifying products at high risk of being discarded based on demand forecasts,

[0888] A means of proposing sales promotion measures for identified products,

[0889] A means of providing discount information on products in accordance with the estimated sentiment of consumers,

[0890] The system according to claim 1, including the following:

[0891] (Claim 3)

[0892] Evaluate the difference between actual sales data and forecast data.

[0893] Methods for retraining AI models to improve the accuracy of demand forecasting,

[0894] A means of adjusting sales strategies by utilizing estimated sentiment information,

[0895] The system according to claim 1, including the following: [Explanation of symbols]

[0896] 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. Means of obtaining sales history information, A means of predicting demand based on acquired sales history information, A means for generating an optimal ordering plan based on predicted demand, A means for displaying the generated order plan, A system that includes this.

2. A means of identifying products at high risk of being discarded based on demand forecasts, A means of proposing sales promotion measures for identified products, The system according to claim 1, including the following:

3. Evaluate the difference between actual sales data and forecast data. Methods for retraining AI models to improve the accuracy of demand forecasting, The system according to claim 1, including the following:

Images