System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting data; means for pre-processing the collected data; means for analyzing with a generative AI model using the pre-processed data; means for generating a recommendation based on results of the analysis; means for visually displaying the recommendation; and means for capturing new data and retraining the generative AI model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot 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] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Inaccurate demand forecasts, poor inventory management, and an inability to adapt to customer preferences and regional trends are problems in the food supply chain. These problems lead to excess inventory, shortages, and poor customer satisfaction, negatively impacting sales and profits. Traditional systems have struggled to effectively solve these problems on a single platform. [Means for solving the problem]

[0005] To solve this problem, the present invention provides a system including a means for collecting data, a means for preprocessing the collected data, a means for analyzing the preprocessed data with a generative AI model, a means for generating recommendations based on the analysis results, a means for visually displaying the recommendations, and a means for incorporating new data and retraining the generative AI model.

[0006] The system of the present invention collects sales data, inventory data, customer purchase history data, and customer order history data from retail and food service businesses, and also collects trend data from external sources. It also includes preprocessing means for filling in missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. This makes it possible to optimize the balance between supply and demand and automatically generate proposals tailored to customer preferences and regional trends.

[0007] "Means of collecting data" refers to a system for electronically obtaining information such as sales, inventory, and customer purchasing history from retail and food service businesses such as supermarkets and restaurants.

[0008] "Preprocessing means" refers to a mechanism for preparing collected data for analysis by completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[0009] A "generative AI model" is an algorithm that uses technologies such as machine learning and deep learning to analyze data and predict future demand and analyze customer preferences.

[0010] The "means of analysis" refers to a mechanism that analyzes data preprocessed by a generative AI model and processes it to derive demand forecasts, customer preferences, and trends.

[0011] The "means of generating proposals" is a mechanism that formulates and presents specific product order quantities, sales promotion measures, inventory management methods, etc. based on the analysis results.

[0012] "Visual display means" refers to a system that displays the generated proposals as graphs or lists in an easy-to-read format for store managers and staff.

[0013] "Retraining" is the process of updating and optimizing generative AI models every time new data is collected. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 1, a 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.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] 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 of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 2, in the data processing device 12, a specific process 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" according to the technology of the present 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 process 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.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative AI techniques for data collection, pre-processing, analysis, recommendation generation, and visual display.

[0036] Program processing flow

[0037] 1. Data Collection

[0038] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants. It also takes into account the influence of external factors by simultaneously obtaining social media and local trend data.

[0039] 2. Data Preprocessing

[0040] The server performs various preprocessing operations on the collected data, such as filling in missing values, detecting and correcting outliers, standardizing data, and removing duplicate data. For example, if there are gaps in sales data, the server fills in the estimated value based on other related data.

[0041] 3. Data analysis

[0042] The server inputs the preprocessed data into a generative AI model for detailed analysis. This includes demand forecasting and customer preference analysis. For example, it predicts next week's demand based on past data and extracts patterns of products preferred by specific customers. It also analyzes regional trends and predicts the popularity of specific products.

[0043] 4. Proposal generation

[0044] The server generates specific recommendations based on the analysis results of the generative AI model, such as next week's ordering list and inventory management measures for a supermarket, or new menu development and promotional campaigns for a restaurant.

[0045] 5. Display results

[0046] The terminals provide managers and staff with visual suggestions, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can view next week's order list on the terminal and make any necessary adjustments.

[0047] 6. Continuous learning

[0048] The server retrains the generative AI model as new data is collected, allowing the system to become more accurate over time and provide increasingly optimal recommendations.

[0049] Specific examples

[0050] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[0051] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[0052] In this way, the system enables retailers and food service businesses to efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[0053] The processing flow will be explained below.

[0054] Program processing steps

[0055] Step 1: Data collection

[0056] The server obtains sales data, inventory data, customer purchase history data, and customer order history data from each retail and restaurant store via API. It also simultaneously collects social media and regional trend data and stores it as external environment data.

[0057] Step 2: Data Preprocessing

[0058] The server performs preprocessing on the collected data. Specifically, it performs operations such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, it fills in estimated values ​​based on similar data from the past.

[0059] Step 3: Demand forecast

[0060] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[0061] Step 4: Customer preference analysis

[0062] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[0063] Step 5: Trend analysis

[0064] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[0065] Step 6: Proposal Generation

[0066] Based on the analysis results, the server generates proposals for each store, such as the optimal product list, order quantities, and sales promotion measures. For example, it generates next week's order list and inventory management measures for a supermarket, and new menu items and sales promotion campaigns for a restaurant.

[0067] Step 7: View the results

[0068] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, and campaign proposals. Managers can review the display and make adjustments as necessary.

[0069] Step 8: Take action

[0070] Users (store managers and staff) follow instructions on the terminal to place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and planning and implementing sales promotion campaigns.

[0071] Step 9: Continuous learning

[0072] The server retrains the generative AI model based on new data collected, thereby maintaining the accuracy of the predictive algorithm and improving the system's performance over time.

[0073] These are the specific processing steps in the AI ​​Food Partner program, which will optimize the food supply chain, improve efficiency in the retail and restaurant industries, and increase customer satisfaction.

[0074] Example 1

[0075] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0076] Managing food supply chains in the retail and restaurant industries requires efficient collection and analysis of a wide range of data and the generation of optimal proposals. However, conventional methods require a great deal of time and effort for data collection and preprocessing, and the accuracy of the analysis is limited. As a result, proper inventory management and demand forecasting are difficult, resulting in excess inventory and lost opportunities. The present invention aims to provide a system that solves these problems.

[0077] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0078] In this invention, the server includes a means for collecting data, a means for preprocessing the collected data, and a means for analyzing the preprocessed data with a generative artificial intelligence model. This allows for automatic collection of data from multiple data sources and highly accurate preprocessing and analysis, thereby enabling efficient inventory management and demand forecasting.

[0079] "Means for collecting data" refers to communication modules and software for automatically acquiring various data, including sales data, inventory data, customer purchase history data, and order history data, from each retail and food service store.

[0080] The "means for preprocessing the collected data" refers to algorithms or software modules for completing missing values ​​in the collected data, detecting and correcting outliers, standardizing data formats, and removing duplicate data.

[0081] "Means of analysis using generative artificial intelligence models" refers to machine learning models and algorithms that use preprocessed data to perform detailed analysis such as demand forecasting and customer preference analysis.

[0082] The "means for generating proposals" refers to algorithms or software modules that create specific action plans, such as inventory management measures, new menu proposals, and sales promotion campaigns, based on the analysis results of the generative artificial intelligence model.

[0083] "Means for visually displaying proposals" refers to a dashboard or user interface that displays the generated proposals and analysis results in graphs or tables in a way that is easy for managers and staff to understand.

[0084] "Means for incorporating new data and retraining the generated AI model" refers to algorithms or software that use new data collected periodically to retrain the AI ​​model to improve its accuracy.

[0085] "Retail sales data" refers to data that includes transaction information when a product is sold at a retail store.

[0086] "Inventory data" refers to data that includes information indicating the inventory status of a retail store or restaurant at a specific point in time.

[0087] "Customer purchase history data" is data that includes records of products and services purchased by individual customers in the past.

[0088] "Customer order history data" refers to data in the food and beverage industry that includes records of food and drink orders that customers have made in the past.

[0089] "Trend data" is external data that shows consumption trends and fashions in a specific region or period, and is obtained from social media and local information services.

[0090] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative artificial intelligence techniques to collect, preprocess, analyze, generate recommendations, and visually display data. The details of this system are described below.

[0091] The server uses APIs and communication modules to collect data from each retail and restaurant chain. For example, it obtains sales and inventory data from a supermarket's sales database, and customer order history from a restaurant's ordering system. It also collects social media data and local trend data in parallel, taking into account the influence of external factors.

[0092] The collected data is preprocessed by the server. This preprocessing includes filling in missing data, correcting inventory data, standardizing data formats, and deleting duplicate data. Specifically, if there are missing values ​​in the sales data, they are filled in based on other related data. Anomaly detection is performed using TENSORFLOW (registered trademark) and PyTorch algorithms to maintain data quality.

[0093] The server then inputs the preprocessed data into generative artificial intelligence models to perform demand forecasting and customer preference analysis. This analysis uses natural language processing (NLP) models and machine learning algorithms. For example, a model is run to predict next week's demand based on past sales data and local weather data. Another model is run to extract customer purchasing patterns and predict the popularity of certain products.

[0094] Based on the analysis results, the server generates specific proposals, such as next week's order list and inventory management for supermarkets, and new menu development and promotion campaigns for restaurants, to ensure smooth supply chain management and maximize sales opportunities.

[0095] The terminals display visual suggestions to managers and staff, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can see next week's order list displayed on the terminal and make necessary adjustments accordingly.

[0096] The system improves its accuracy through continuous learning: as new data is collected, the server retrains the generative artificial intelligence model to provide more optimal recommendations, allowing the system to evolve over time and become even more effective.

[0097] Examples of specific prompts for generative AI models

[0098] The following sentence can be used as an example of a prompt sentence to be input to the generative artificial intelligence model:

[0099] Example prompt 1:

[0100] "Predict next week's demand for ice cream based on weekly sales data from the past year and local weather data."

[0101] Example prompt 2:

[0102] "Using customer purchase history and social media trend data from the past three months, please suggest new menu items that will be most popular next month."

[0103] This makes it possible to automatically generate specific and effective proposals for the retail and food service industries.

[0104] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0105] Step 1: Data collection

[0106] The server collects various data from each retail and restaurant chain. Specifically, the server accesses the supermarket's sales database via API to obtain sales and inventory data. It also obtains customer order history from the restaurant's ordering system, and uses social media APIs and local information service APIs to collect social media data and local trend data. The input is the various data received from the API, and the output is raw collected data that is stored in a central database within the server.

[0107] Step 2: Data Preprocessing

[0108] The server preprocesses the collected data. First, if there is missing data, it fills it in based on other relevant data. Next, it uses statistical methods and machine learning algorithms to detect and correct outliers. If the data format is different, it is converted into a consistent format, and if there is duplicate data, it is removed. The input is the raw data collected in step 1, and the output is the clean data after preprocessing.

[0109] Step 3: Data analysis

[0110] The server inputs the preprocessed data into a generative artificial intelligence model for further analysis. For example, it uses TensorFlow and PyTorch to run demand forecasting and customer preference analysis models. Specifically, it uses past sales data and weather data to predict future demand and analyze customer purchasing patterns. The input is the preprocessed data, and the output is the analysis results, which include predicted sales volume, a list of popular products, and trend analysis results.

[0111] Step 4: Proposal Generation

[0112] The server generates specific proposals based on the analysis results generated. For example, it provides a supermarket with an order list for the next week based on the results of demand forecasting, or creates a promotional campaign for a specific customer group based on the results of customer preference analysis. It also proposes new menu additions to restaurants based on regional trends. The input is the analysis results generated in step 3, and the output is specific proposals.

[0113] Step 5: View the results

[0114] The terminal visualizes and displays the proposals provided by the server. It displays sales forecasts in graph form, order lists in list form, and campaign proposals on a dashboard accessible to managers and staff. For example, a supermarket manager can check next week's order list on the terminal and make necessary adjustments. The input is the proposal generated in step 4, and the output is the visual representation of the proposal.

[0115] Step 6: Continuous learning

[0116] The server retrains the generative AI model using newly collected data, which improves the model's accuracy over time and enables it to provide more accurate and useful suggestions. The input is the latest newly collected data, and the output is a retrained, highly accurate generative AI model.

[0117] (Application example 1)

[0118] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0119] Traditional food supply chain management systems are limited to collecting and analyzing data within stores and warehouses, and do not optimize the delivery process. This makes it difficult to select efficient routes during delivery or monitor progress in real time, which can lead to placement losses and deterioration of storage quality. Furthermore, delays in demand forecasting and response, as well as difficulties in optimal ordering and inventory management, lead to lost sales opportunities and excess inventory.

[0120] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0121] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating proposals based on the analysis results, means for visually displaying the proposals, means for incorporating new data and retraining the generative AI model, and means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles. This enables real-time data collection and analysis during the delivery process, making it possible to propose optimal delivery routes and times, and to realize efficient ordering and inventory management based on demand forecasts.

[0122] "Data collection means" refers to means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, and customer purchase history data from retail and food service businesses.

[0123] The "preprocessing means" refers to means for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[0124] A "generative AI model analysis means" is a means for analyzing pre-processed data using a generative AI model to forecast demand, customer preferences, optimize delivery routes, and predict delivery times.

[0125] The "proposal generation means" is a means for generating optimal delivery routes, delivery times, and order lists based on the analysis results of the generative AI model.

[0126] The "visual display means" is a means for displaying generated proposals and analysis results in a format that is easy for managers and staff to understand.

[0127] "Retraining methods" are methods for incorporating new collected data into a generative AI model and retraining it to improve the model's accuracy.

[0128] An "autonomous vehicle" is a vehicle that can drive autonomously and deliver goods to a destination.

[0129] This invention is a food supply chain management system that uses autonomous vehicles, and aims to achieve efficient and effective data collection, data preprocessing, analysis using generative AI models, proposal generation, and visual display of results.

[0130] First, the server collects data. Specifically, it obtains location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, customer purchase history data, and order history data from partner stores and restaurants. It also simultaneously obtains trend data from external sources such as social media.

[0131] The server then preprocesses the collected data. This includes filling in missing values, detecting and correcting outliers, standardizing data formats, and removing duplicate data. For example, if there are missing sales data, a method is used to fill in estimated values ​​using other related data.

[0132] The preprocessed data is passed to a generative AI model. The server uses the generative AI model to perform detailed analysis of the data. Specific analysis items include demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. This makes it possible to predict demand for the following week based on past data, or predict the popularity of specific products.

[0133] Based on the analysis results, the server generates specific proposals, such as routes and delivery schedules for autonomous vehicles, next week's order lists and inventory management strategies for supermarkets and restaurants, and even suggestions for developing new menus and promotional campaigns.

[0134] The generated proposals are visually displayed on the terminal. The proposals are displayed in graph and list format on the terminal, making them easy for managers and staff to understand and implement. For example, a supermarket manager can check the order list on the terminal and make any necessary adjustments.

[0135] The server retrains the generative AI model every time new data is collected, improving the accuracy of the system, which allows it to provide timely and more optimal recommendations.

[0136] Specific examples

[0137] For example, if the server predicts that demand for ice cream will increase in a particular region during the summer, it will combine past sales data with local weather data to predict ice cream demand for the following week. As a result, a suggestion to order twice the usual amount will be displayed on the terminal. It may also analyze customer purchase history data to determine that a particular ice cream (such as chocolate) is very popular, and suggest launching a promotional campaign using this as a main product.

[0138] Example prompt sentence:

[0139] "Design an algorithm to predict ice cream demand in a specific region during the summer and suggest the optimal order quantity."

[0140] "Consider sales data from the past three months to forecast next week's sales of essential items and generate an optimal order list."

[0141] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0142] Step 1:

[0143] The server collects data, specifically location information from autonomous vehicles, delivery progress data, and warehouse temperature data, as well as sales data, inventory data, customer purchase history data, order history data, and external social media trend data from partner stores and restaurants. This allows the current status of the entire supply chain to be collected as data.

[0144] Input: location information from self-driving vehicles, delivery progress data, warehouse temperature data, sales data from stores and restaurants, inventory data, purchase history data, order history data, SNS trend data

[0145] Output: Raw data collected

[0146] Step 2:

[0147] The server performs preprocessing on the collected data, including filling in missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are missing sales data, it fills in the estimated value based on other related data. This ensures data consistency and quality.

[0148] Input: Raw data collected

[0149] Output: Preprocessed data

[0150] Step 3:

[0151] The server inputs the preprocessed data into a generative AI model for detailed analysis. Specifically, it performs demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. The generative AI model predicts future trends from past data and suggests optimal actions.

[0152] Input: Preprocessed data

[0153] Output: Analysis results

[0154] Step 4:

[0155] The server generates specific proposals based on the analysis results of the generative AI model, such as optimal delivery routes and schedules for autonomous vehicles, order lists and inventory management strategies for the following week, and proposals for the development of new menu items and sales promotion campaigns.

[0156] Input: Analysis results

[0157] Output: Specific proposals (optimal delivery routes and schedules, order lists, inventory management measures, proposals for developing new menu items, proposals for sales promotion campaigns)

[0158] Step 5:

[0159] The server visually displays the generated proposals on the terminal, where they are displayed in graph and list format, providing managers and staff with a format that is easy to understand. This allows managers and staff to intuitively understand the proposals and make any necessary adjustments or actions.

[0160] Input: Specific proposal

[0161] Output: Visual display (graphs, list of suggestions)

[0162] Step 6:

[0163] The server retrains the generative AI model every time new data is collected, improving the model's accuracy and its ability to provide better recommendations. The retrained model ensures that predictions and recommendations are always based on the most up-to-date data.

[0164] Input: New data

[0165] Output: Retrained generative AI model

[0166] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0167] This invention relates to a system that combines generative AI technology with an emotion engine that recognizes user emotions to efficiently manage food supply chains in the retail and restaurant industries. The system provides functions for data collection, preprocessing, analysis, recommendation generation, visual display, continuous learning, and emotion recognition.

[0168] Program processing flow

[0169] 1. Data Collection

[0170] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants via API. It also simultaneously collects social media and local trend data and stores it as external environment data.

[0171] 2. Data Preprocessing

[0172] The server performs various preprocessing operations on the collected data, such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, the server fills in estimated values ​​based on similar data from the past.

[0173] 3. Demand forecasting

[0174] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[0175] 4. Customer preference analysis

[0176] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[0177] 5. Trend Analysis

[0178] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[0179] 6. Emotion recognition

[0180] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to determine whether they are expressing positive or negative emotions. For example, if a customer's posts are positive, it can suggest new product information or promotions to the customer.

[0181] 7. Emotion-based suggestion generation

[0182] The server generates customized marketing messages and new product suggestions based on the customer's emotional state analyzed by the emotion engine, and if the customer is dissatisfied, suggests special offers or services to resolve the problem.

[0183] 8. Results display

[0184] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can review the display and make adjustments as necessary.

[0185] 9. Continuous learning

[0186] The server retrains the generative AI model and emotion engine as new data is collected, allowing the system to become more accurate over time and capable of providing more optimal recommendations.

[0187] Specific examples

[0188] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[0189] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[0190] Furthermore, the emotion engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the customer an invitation to a tasting event for the new menu. The suggestion is displayed on the terminal, and the restaurant manager sends the invitation message to the customer according to the suggestion.

[0191] In this way, by combining generative AI technology with emotion recognition, the system can help retail and food service businesses efficiently and effectively manage their food supply chains, increasing customer satisfaction and sales.

[0192] The processing flow will be explained below.

[0193] Program processing steps

[0194] Step 1: Data collection

[0195] The server collects various data from each retail and restaurant establishment. Specifically, it periodically obtains sales data, inventory data, customer purchase history data, and customer order history data via API. At the same time, it also collects social media and regional trend data and stores it as external environment data for analysis.

[0196] Step 2: Data Preprocessing

[0197] The server performs preprocessing on the collected data. During this process, it complements missing values ​​in the data, detects and corrects outliers, standardizes data formats, and removes duplicate data. For example, if there is a defect in sales data, it can infer the defect from other related data and complement it.

[0198] Step 3: Demand forecast

[0199] The server then inputs the pre-processed data into a generative AI model to forecast demand for the next week or month. Based on past sales data, inventory data, and regional trend data, the server makes detailed forecasts of demand for specific products and days.

[0200] Step 4: Customer preference analysis

[0201] The server analyzes customer purchase and order history data to identify which products and categories a particular customer prefers, and uses this information to generate personalized marketing messages promoting specific products and services.

[0202] Step 5: Trend analysis

[0203] The server analyzes social media data and regional trend data to identify products and categories that are likely to be popular now and in the near future. Based on this information, it makes suggestions for adding new products or running promotional campaigns, for example.

[0204] Step 6: Emotion Recognition

[0205] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to identify positive and negative emotions. For example, if a customer has positive feelings about a new product, that emotion can be used to promote the new product.

[0206] Step 7: Sentiment-based suggestion generation

[0207] The server generates customized marketing messages and new product proposals based on the customer's emotional state analyzed by the emotion engine. If a customer is dissatisfied with a particular product or service, the server will suggest special offers or services to resolve the issue.

[0208] Step 8: View the results

[0209] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can check the display and adjust and implement the proposals as necessary.

[0210] Step 9: Take Action

[0211] Users (store managers and staff) follow instructions on the terminal to actually place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and sending customized messages to specific customers based on the results of sentiment analysis.

[0212] Step 10: Continuous learning

[0213] The server retrains the generative AI model and sentiment engine based on new data collected, further improving accuracy over time, giving the system the ability to provide increasingly accurate suggestions.

[0214] These are the specific processing steps in the AI ​​Food Partner system that combines an emotion engine. This is expected to improve the efficiency of the food supply chain, improve customer satisfaction, and increase sales.

[0215] Example 2

[0216] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0217] Efficient management of food supply chains in the retail and restaurant industries requires the collection and analysis of a wide variety of data, and marketing that takes customer emotions into consideration is also required. Conventional systems have difficulty processing this data in a centralized manner, which can lead to insufficient demand forecasting and analysis of customer preferences. Furthermore, it is difficult to generate proposals that take customer emotions into consideration, leaving challenges in improving customer satisfaction and sales.

[0218] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating suggestions based on the analysis results, means for visually displaying the suggestions, means for incorporating new data and retraining the generative AI model, means for analyzing emotions using an emotion engine that recognizes the emotional state of customers, and means for generating customized suggestions based on the analyzed emotional state. This enables efficient management of the food supply chain and improves customer satisfaction and sales by providing suggestions based on customer preferences and emotions.

[0219] The "means of collecting data" refers to a mechanism for collecting sales data, inventory data, customer purchase history data, order history data, and external trend data and social media data from each retail and food service store.

[0220] The "preprocessing means" is a processing mechanism for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[0221] A "means for analyzing using a generative AI model" is a mechanism that uses preprocessed data as input and analyzes and predicts the data using a specific algorithm.

[0222] The "means for generating proposals" is a mechanism for creating customized marketing messages and product proposals based on the analysis results and sentiment analysis results.

[0223] The "visual display means" refers to a display device and its control mechanism for visually presenting the generated proposals and analysis results to managers and staff.

[0224] A "means for retraining" is a mechanism for retraining a generative AI model to improve its accuracy each time new data is acquired.

[0225] An "emotion engine" is an analytical device that analyzes a customer's emotional state from text and other data and determines whether the emotion is positive, negative, or neutral.

[0226] The "means for generating customized offers" is a mechanism for generating offers and marketing messages that are suited to individual customers based on the analyzed emotional state.

[0227] MODE FOR CARRYING OUT THE INVENTION

[0228] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries, improving customer satisfaction and sales. The system provides functions for data collection, data preprocessing, analysis using generative AI models, recommendation generation, visual display, continuous learning, emotion recognition, and customized recommendations.

[0229] Hardware and software used

[0230] server

[0231] The server is a computer that runs the following software:

[0232] Data collection software: Acquires data via API, collects data using web crawling technology, etc.

[0233] Data preprocessing software: Using data analysis tools such as Pandas and Scikit-learn, we perform tasks such as imputing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[0234] Generative AI models: Data analysis and demand forecasting, including time series forecasting models built using PyTorch and TensorFlow.

[0235] Emotion Recognition Engine: Analyze customer emotions using an emotion analysis engine such as the Google® Cloud Natural Language API.

[0236] Retraining capabilities: Ability to retrain generative AI models and sentiment analysis engines with each information update.

[0237] Terminal

[0238] The terminal is a computer or tablet for running the following software:

[0239] Visual display software: Use libraries such as D3.js and Chart.js to visually display recommendations and analysis results.

[0240] Specific examples

[0241] For example, when a local supermarket needs to forecast demand for a particular product, the server collects past sales data, inventory data, and local trend data (such as past weather information and social media trend information). This data is then preprocessed and input into a generative AI model. If the result predicts that demand next week will be twice the usual amount, the device will visually display the detailed forecast along with a suggestion to double the order quantity.

[0242] Furthermore, when analyzing social media posts using emotion recognition, it can be determined that a particular dish (e.g., homemade pizza) is popular in a region. Based on this information, the server generates a proposal to add homemade pizza to the menu and run a campaign on social media. The proposal and specific campaign methods are visually displayed on the device.

[0243] Furthermore, the emotion recognition engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the relevant customer an invitation message to a tasting event for the new menu. The suggestion is displayed on the device, and the user can send the message to the customer according to the suggestion.

[0244] Prompt Sentence Examples

[0245] Demand forecast prompt statement

[0246] python

[0247] past_sales_data = get_past_sales_data()

[0248] trend_data = get_trend_data()

[0249] input_data = combine_data(past_sales_data, trend_data)

[0250] predicted_demand = ai_model.predict(input_data)

[0251] print(predicted_demand)

[0252] Customer Preference Analysis Prompt

[0253] python

[0254] customer_purchase_history = get_customer_purchase_history(customer_id)

[0255] recommendations = recommend_products(customer_purchase_history)

[0256] print(recommendations)

[0257] As described above, by combining generative AI technology with emotion recognition, this invention can help retail and food service businesses efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[0258] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0259] Program processing flow and specific explanation

[0260] Step 1: Data collection

[0261] The server collects data from each retail and restaurant establishment, including the following data:

[0262] Input: Sales data, inventory data, customer purchase history, order history, social media data, and regional trend data obtained using API and web crawling technology.

[0263] Data processing: Store data is obtained in real time through APIs, and social media and trend data is collected by web crawling.

[0264] Output: Various datasets stored in a database.

[0265] Specifically, the server retrieves supermarket sales data through an API call and executes a crawling script to collect social media trends.

[0266] Step 2: Data Preprocessing

[0267] The server performs pre-processing on the collected data.

[0268] Input: Raw data collected in step 1.

[0269] Data processing: Using Pandas and Scikit-learn, we perform missing value imputation, outlier detection and correction, data format standardization, and duplicate data removal.

[0270] Output: A preprocessed and clean dataset.

[0271] As a specific operation, for example, if there is a missing value in the sales data, the server will calculate the average value from similar past data and fill in the missing value.

[0272] Step 3: Demand forecast

[0273] The server uses a generative AI model to predict demand.

[0274] Input: The preprocessed dataset from step 2.

[0275] Data Computing: Use time series forecasting models to forecast demand for the next week. Use generative AI models in PyTorch and TensorFlow.

[0276] Output: Forecasted demand data.

[0277] Specifically, the server inputs past sales and inventory data into the generative AI model, obtaining a prediction result such as, "Demand for ice cream will be twice the usual amount next Tuesday."

[0278] Step 4: Customer preference analysis

[0279] The server analyzes the customer's purchasing history and other information to identify individual preferences.

[0280] Input: Customer purchase and order history data preprocessed in step 2.

[0281] Data computation: Using Python and Scikit-learn, data is clustered to identify customer preferences.

[0282] Output: A list of each customer's preferred products and categories.

[0283] Specifically, the server may identify, for example, that a customer has frequently purchased chocolate-related products over the past six months, and then use that information to suggest a special campaign.

[0284] Step 5: Trend analysis

[0285] The server analyzes social media data and regional trend data to identify trends.

[0286] Input: Social media data and local trend data collected in Step 1.

[0287] Data Computing: Use NLTK and Spacy to perform text analysis and identify trending products and categories.

[0288] Output: A list of trending items and categories.

[0289] Specifically, the server detects trends such as "homemade pizza is becoming a hot topic on local social media."

[0290] Step 6: Emotion Recognition

[0291] The server analyzes the customer's emotions using an emotion engine.

[0292] Input: Text data such as social media posts and reviews.

[0293] Data calculation: Uses Google Cloud Natural Language API to identify positive, negative, and neutral sentiment.

[0294] Output: Customer sentiment analysis data.

[0295] As a specific operation, the server obtains a result such as "There are many posts expressing very positive sentiment toward this new dessert menu."

[0296] Step 7: Sentiment-based suggestion generation

[0297] The server generates customized suggestions based on the emotional state analyzed by the emotion engine.

[0298] Input: Sentiment analysis result data from step 6.

[0299] Data Computing: Generate customized marketing messages and new product offers using Python and JavaScript®.

[0300] Output: A customized suggestion message.

[0301] As a specific example of its operation, the server may make a suggestion such as "generating an invitation message to a tasting event of a new menu for customers who have expressed positive emotions."

[0302] Step 8: View the results

[0303] The terminal visually displays the suggestions received from the server to the manager or staff.

[0304] Input: Proposed data and analysis results received from the server.

[0305] Display: Display in graph or list format using D3.js or Chart.js.

[0306] Output: Visually displayed recommendations and analysis results.

[0307] Specifically, the device will display a message with a graph, such as, "Demand for ice cream will be twice the usual amount next Tuesday, so please double your order quantity."

[0308] Step 9: Continuous learning

[0309] The server takes in new data and retrains the generative AI model and emotion engine.

[0310] Input: The newly collected dataset.

[0311] Data computation: Update model parameters and retrain using TensorFlow or PyTorch.

[0312] Output: An improved generative AI model and emotion engine.

[0313] Specifically, as data is updated weekly, sales trends and changes in customer sentiment are reflected in the model, improving the accuracy of the next forecast.

[0314] (Application example 2)

[0315] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0316] The retail and restaurant industries require systems that efficiently manage food supply chains while taking into account customer emotions. However, current systems are unable to generate proposals that fully reflect customer emotions and preferences, posing challenges in improving customer satisfaction and making appropriate demand forecasts. Therefore, a method is needed that analyzes customers' emotional states using a variety of data and makes personalized proposals.

[0317] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0318] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating recommendations based on the analysis results, means for visually displaying the recommendations, means for incorporating new data and retraining the generative AI model, means for analyzing user sentiment, and means for generating recommendations based on the sentiment analysis results, thereby enabling personalized food supply chain management that takes into account the emotional state of customers.

[0319] "Means for collecting data" refers to a mechanism for obtaining sales data, inventory data, customer purchasing history data, social media data, and regional trend data from each retail and food service store.

[0320] The "preprocessing means" is a mechanism for complementing missing values ​​in collected data, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[0321] "Means for analysis using a generative AI model" refers to a mechanism for performing demand forecasting and customer preference analysis using a generative AI model based on preprocessed data.

[0322] The "means for generating proposals" is a mechanism for creating personalized marketing messages and new product proposals based on the analysis results and sentiment analysis results of the generative AI model.

[0323] The "visual display means" is a mechanism for visually displaying the proposals received from the server to the administrator or staff in the form of graphs, lists, etc.

[0324] A "means for retraining the generative AI model" is a mechanism for retraining the generative AI model each time new data is collected, improving its accuracy over time.

[0325] "Means for analyzing user emotions" refers to a mechanism that analyzes customer purchasing behavior and social media posting data to determine whether emotions are positive or negative.

[0326] The "means for generating proposals based on the results of sentiment analysis" is a mechanism for generating customized marketing messages and new product proposals for customers based on the results of sentiment analysis.

[0327] "Means for collecting retail sales data, inventory data, and customer purchase history data" refers to mechanisms such as APIs and sensors for obtaining sales, inventory, and customer purchase history from retail stores.

[0328] "Means for collecting food service sales data and customer order history data" refers to mechanisms such as APIs and sensors for obtaining sales and specific customer order histories from food service establishments.

[0329] "Means for collecting trend data from outside" refers to a mechanism for obtaining regional trend information from social media and the Internet.

[0330] The "means for imputing missing values ​​and detecting and correcting outliers" is a mechanism for estimating and imputing missing values, detecting outliers, and appropriately correcting them for collected data.

[0331] The "means for unifying data formats and deleting duplicate data" is a mechanism for unifying data of different formats and deleting duplicate data.

[0332] The "means for generating personalized proposals based on the results of sentiment analysis" is a mechanism for using the results of sentiment analysis to make personalized product and service proposals to individual customers.

[0333] MODE FOR CARRYING OUT THE INVENTION

[0334] This invention is a system for efficiently and effectively managing food supply chains in the retail and restaurant industries. The system combines generative AI technology with an emotion engine that recognizes user emotions to provide data collection, preprocessing, analysis, recommendation generation, visual display, and continuous learning functions. The system can also be implemented as a smartphone application. Specific embodiments of the system are described below.

[0335] Hardware and Software

[0336] Hardware

[0337] Smartphone (iOS or ANDROID (registered trademark))

[0338] Software Framework

[0339] requests: for API calls

[0340] json: For working with JSON data

[0341] textblob: for sentiment analysis

[0342] sklearn: For data preprocessing and demand forecasting

[0343] pandas: for data preprocessing

[0344] Data flow

[0345] The server processes the data through the following steps:

[0346] 1. Data Collection

[0347] The server collects sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant establishments via API.

[0348] 2. Data Preprocessing

[0349] The server complements missing values ​​in the collected data, detects and corrects outliers, standardizes data formats, and deletes duplicate data.

[0350] 3. Demand forecasting

[0351] The server inputs the pre-processed data into a generative AI model to predict future demand.

[0352] 4. Emotion recognition

[0353] The server uses an emotion engine to analyze the user's emotional state and determine whether the emotion is positive or negative based on social media posts and purchase history data.

[0354] 5. Proposal generation

[0355] The server generates personalized marketing messages and new product suggestions based on the analysis results of the generative AI model and the results of sentiment analysis.

[0356] 6. Results display

[0357] The terminal visually displays the suggestions received from the server to managers and staff in the form of graphs and lists.

[0358] 7. Continuous learning

[0359] The server retrains the generative AI model and emotion engine every time new data is collected.

[0360] Specific examples

[0361] For example, suppose a user enters the text "I'm feeling really down today..." To parse this text, the server does the following:

[0362] 1. Emotion recognition

[0363] The server performs sentiment analysis using the TextBlob library and returns a negative sentiment score.

[0364] 2. Proposal generation

[0365] The server generates a suggestion, "Comfort food specially for you!" based on the negative sentiment score, past purchase history, and trend data.

[0366] The suggestions are visually displayed to the user in an app on their device, providing them with food delivery suggestions tailored to their emotional state that day.

[0367] Prompt Sentence Examples

[0368] An example of a prompt sentence to input to the generative AI model is:

[0369] "Create recommended food delivery suggestions based on the user's purchasing history and sentiment score. For example, if a user types, 'I am feeling so happy today!', provide suggestions based on positive sentiment."

[0370] Examples include:

[0371] In this way, the system of the present invention improves customer satisfaction and realizes efficient management of the food supply chain by providing personalized suggestions based on the user's emotional state.

[0372] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0373] Step 1:

[0374] Data collection

[0375] The server uses APIs to obtain sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant stores. This data is collected in JSON format. The input is the store's API key, and the output is a JSON-formatted dataset.

[0376] Step 2:

[0377] Data Preprocessing

[0378] The server performs missing value completion, outlier detection and correction, data format standardization, and duplicate data removal for the collected data. The input is the JSON dataset obtained in step 1. For each data entry, if a missing value exists, it is completed based on similar data, and outliers are corrected according to certain criteria. The output is a clean dataset.

[0379] Step 3:

[0380] Demand forecasting

[0381] The server inputs the preprocessed data into a generative AI model to predict future demand. The input is the preprocessed dataset, and the output is the demand forecast data for the next week. Specifically, it uses past sales data, inventory data, and regional trend data to predict demand for a specific day or specific product.

[0382] Step 4:

[0383] emotion recognition

[0384] The server uses an emotion engine to analyze the user's emotional state. It determines whether the emotion is positive or negative by analyzing social media posts and purchase history data. The input is social media posts and purchase history data, and the output is an emotion score.

[0385] Step 5:

[0386] Proposal generation

[0387] The server generates personalized marketing messages and new product proposals based on the analysis results of the generative AI model and the sentiment analysis results. The input is demand forecast data and sentiment scores, and the output is personalized proposals. Specifically, it prepares proposals with rewards for positive sentiment scores and includes consolation messages for negative sentiment scores.

[0388] Step 6:

[0389] Results display

[0390] The terminal visually displays the proposals received from the server in graph or list format to managers and staff. The input is the proposal data from the server, and the output is the visually displayed information. Specifically, it displays campaign proposals and customer support methods on a dashboard for managers.

[0391] Step 7:

[0392] Continuous Learning

[0393] The server retrains the generative AI model and emotion engine every time new data is collected. The input is a new dataset, and the output is a retrained model. Specifically, the model is periodically updated to improve its accuracy.

[0394] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the 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.

[0395] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0396] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0397] [Second embodiment]

[0398] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0399] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0400] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0401] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0402] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0403] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0404] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0405] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0406] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0407] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0408] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0409] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0410] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative AI techniques for data collection, pre-processing, analysis, recommendation generation, and visual display.

[0411] Program processing flow

[0412] 1. Data Collection

[0413] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants. It also takes into account the influence of external factors by simultaneously obtaining social media and local trend data.

[0414] 2. Data Preprocessing

[0415] The server performs various preprocessing operations on the collected data, such as filling in missing values, detecting and correcting outliers, standardizing data, and removing duplicate data. For example, if there are gaps in sales data, the server fills in the estimated value based on other related data.

[0416] 3. Data analysis

[0417] The server inputs the preprocessed data into a generative AI model for detailed analysis. This includes demand forecasting and customer preference analysis. For example, it predicts next week's demand based on past data and extracts patterns of products preferred by specific customers. It also analyzes regional trends and predicts the popularity of specific products.

[0418] 4. Proposal generation

[0419] The server generates specific recommendations based on the analysis results of the generative AI model, such as next week's ordering list and inventory management measures for a supermarket, or new menu development and promotional campaigns for a restaurant.

[0420] 5. Display results

[0421] The terminals provide managers and staff with visual suggestions, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can view next week's order list on the terminal and make any necessary adjustments.

[0422] 6. Continuous learning

[0423] The server retrains the generative AI model as new data is collected, allowing the system to become more accurate over time and provide increasingly optimal recommendations.

[0424] Specific examples

[0425] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[0426] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[0427] In this way, the system enables retailers and food service businesses to efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[0428] The processing flow will be explained below.

[0429] Program processing steps

[0430] Step 1: Data collection

[0431] The server obtains sales data, inventory data, customer purchase history data, and customer order history data from each retail and restaurant store via API. It also simultaneously collects social media and regional trend data and stores it as external environment data.

[0432] Step 2: Data Preprocessing

[0433] The server performs preprocessing on the collected data. Specifically, it performs operations such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, it fills in estimated values ​​based on similar data from the past.

[0434] Step 3: Demand forecast

[0435] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[0436] Step 4: Customer preference analysis

[0437] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[0438] Step 5: Trend analysis

[0439] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[0440] Step 6: Proposal Generation

[0441] Based on the analysis results, the server generates proposals for each store, such as the optimal product list, order quantities, and sales promotion measures. For example, it generates next week's order list and inventory management measures for a supermarket, and new menu items and sales promotion campaigns for a restaurant.

[0442] Step 7: View the results

[0443] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, and campaign proposals. Managers can review the display and make adjustments as necessary.

[0444] Step 8: Take action

[0445] Users (store managers and staff) follow instructions on the terminal to place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and planning and implementing sales promotion campaigns.

[0446] Step 9: Continuous learning

[0447] The server retrains the generative AI model based on new data collected, thereby maintaining the accuracy of the predictive algorithm and improving the system's performance over time.

[0448] These are the specific processing steps in the AI ​​Food Partner program, which will optimize the food supply chain, improve efficiency in the retail and restaurant industries, and increase customer satisfaction.

[0449] Example 1

[0450] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0451] Managing food supply chains in the retail and restaurant industries requires efficient collection and analysis of a wide range of data and the generation of optimal proposals. However, conventional methods require a great deal of time and effort for data collection and preprocessing, and the accuracy of the analysis is limited. As a result, proper inventory management and demand forecasting are difficult, resulting in excess inventory and lost opportunities. The present invention aims to provide a system that solves these problems.

[0452] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0453] In this invention, the server includes a means for collecting data, a means for preprocessing the collected data, and a means for analyzing the preprocessed data with a generative artificial intelligence model. This allows for automatic collection of data from multiple data sources and highly accurate preprocessing and analysis, thereby enabling efficient inventory management and demand forecasting.

[0454] "Means for collecting data" refers to communication modules and software for automatically acquiring various data, including sales data, inventory data, customer purchase history data, and order history data, from each retail and food service store.

[0455] The "means for preprocessing the collected data" refers to algorithms or software modules for completing missing values ​​in the collected data, detecting and correcting outliers, standardizing data formats, and removing duplicate data.

[0456] "Means of analysis using generative artificial intelligence models" refers to machine learning models and algorithms that use preprocessed data to perform detailed analysis such as demand forecasting and customer preference analysis.

[0457] The "means for generating proposals" refers to algorithms or software modules that create specific action plans, such as inventory management measures, new menu proposals, and sales promotion campaigns, based on the analysis results of the generative artificial intelligence model.

[0458] "Means for visually displaying proposals" refers to a dashboard or user interface that displays the generated proposals and analysis results in graphs or tables in a way that is easy for managers and staff to understand.

[0459] "Means for incorporating new data and retraining the generated AI model" refers to algorithms or software that use new data collected periodically to retrain the AI ​​model to improve its accuracy.

[0460] "Retail sales data" refers to data that includes transaction information when a product is sold at a retail store.

[0461] "Inventory data" refers to data that includes information indicating the inventory status of a retail store or restaurant at a specific point in time.

[0462] "Customer purchase history data" is data that includes records of products and services purchased by individual customers in the past.

[0463] "Customer order history data" refers to data in the food and beverage industry that includes records of food and drink orders that customers have made in the past.

[0464] "Trend data" is external data that shows consumption trends and fashions in a specific region or period, and is obtained from social media and local information services.

[0465] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative artificial intelligence techniques to collect, preprocess, analyze, generate recommendations, and visually display data. The details of this system are described below.

[0466] The server uses APIs and communication modules to collect data from each retail and restaurant chain. For example, it obtains sales and inventory data from a supermarket's sales database, and customer order history from a restaurant's ordering system. It also collects social media data and local trend data in parallel, taking into account the influence of external factors.

[0467] The collected data is preprocessed by the server. This preprocessing includes filling in missing data, correcting inventory data, standardizing data formats, and removing duplicate data. Specifically, if there are missing values ​​in sales data, they are filled in based on other related data. Anomaly detection is performed using TensorFlow and PyTorch algorithms to maintain data quality.

[0468] The server then inputs the preprocessed data into generative artificial intelligence models to perform demand forecasting and customer preference analysis. This analysis uses natural language processing (NLP) models and machine learning algorithms. For example, a model is run to predict next week's demand based on past sales data and local weather data. Another model is run to extract customer purchasing patterns and predict the popularity of certain products.

[0469] Based on the analysis results, the server generates specific proposals, such as next week's order list and inventory management for supermarkets, and new menu development and promotion campaigns for restaurants, to ensure smooth supply chain management and maximize sales opportunities.

[0470] The terminals display visual suggestions to managers and staff, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can see next week's order list displayed on the terminal and make necessary adjustments accordingly.

[0471] The system improves its accuracy through continuous learning: as new data is collected, the server retrains the generative artificial intelligence model to provide more optimal recommendations, allowing the system to evolve over time and become even more effective.

[0472] Examples of specific prompts for generative AI models

[0473] The following sentence can be used as an example of a prompt sentence to be input to the generative artificial intelligence model:

[0474] Example prompt 1:

[0475] "Predict next week's demand for ice cream based on weekly sales data from the past year and local weather data."

[0476] Example prompt 2:

[0477] "Using customer purchase history and social media trend data from the past three months, please suggest new menu items that will be most popular next month."

[0478] This makes it possible to automatically generate specific and effective proposals for the retail and food service industries.

[0479] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0480] Step 1: Data collection

[0481] The server collects various data from each retail and restaurant chain. Specifically, the server accesses the supermarket's sales database via API to obtain sales and inventory data. It also obtains customer order history from the restaurant's ordering system, and uses social media APIs and local information service APIs to collect social media data and local trend data. The input is the various data received from the API, and the output is raw collected data that is stored in a central database within the server.

[0482] Step 2: Data Preprocessing

[0483] The server preprocesses the collected data. First, if there is missing data, it fills it in based on other relevant data. Next, it uses statistical methods and machine learning algorithms to detect and correct outliers. If the data format is different, it is converted into a consistent format, and if there is duplicate data, it is removed. The input is the raw data collected in step 1, and the output is the clean data after preprocessing.

[0484] Step 3: Data analysis

[0485] The server inputs the preprocessed data into a generative artificial intelligence model for further analysis. For example, it uses TensorFlow and PyTorch to run demand forecasting and customer preference analysis models. Specifically, it uses past sales data and weather data to predict future demand and analyze customer purchasing patterns. The input is the preprocessed data, and the output is the analysis results, which include predicted sales volume, a list of popular products, and trend analysis results.

[0486] Step 4: Proposal Generation

[0487] The server generates specific proposals based on the analysis results generated. For example, it provides a supermarket with an order list for the next week based on the results of demand forecasting, or creates a promotional campaign for a specific customer group based on the results of customer preference analysis. It also proposes new menu additions to restaurants based on regional trends. The input is the analysis results generated in step 3, and the output is specific proposals.

[0488] Step 5: View the results

[0489] The terminal visualizes and displays the proposals provided by the server. It displays sales forecasts in graph form, order lists in list form, and campaign proposals on a dashboard accessible to managers and staff. For example, a supermarket manager can check next week's order list on the terminal and make necessary adjustments. The input is the proposal generated in step 4, and the output is the visual representation of the proposal.

[0490] Step 6: Continuous learning

[0491] The server retrains the generative AI model using newly collected data, which improves the model's accuracy over time and enables it to provide more accurate and useful suggestions. The input is the latest newly collected data, and the output is a retrained, highly accurate generative AI model.

[0492] (Application example 1)

[0493] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0494] Traditional food supply chain management systems are limited to collecting and analyzing data within stores and warehouses, and do not optimize the delivery process. This makes it difficult to select efficient routes during delivery or monitor progress in real time, which can lead to placement losses and deterioration of storage quality. Furthermore, delays in demand forecasting and response, as well as difficulties in optimal ordering and inventory management, lead to lost sales opportunities and excess inventory.

[0495] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0496] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating proposals based on the analysis results, means for visually displaying the proposals, means for incorporating new data and retraining the generative AI model, and means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles. This enables real-time data collection and analysis during the delivery process, making it possible to propose optimal delivery routes and times, and to realize efficient ordering and inventory management based on demand forecasts.

[0497] "Data collection means" refers to means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, and customer purchase history data from retail and food service businesses.

[0498] The "preprocessing means" refers to means for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[0499] A "generative AI model analysis means" is a means for analyzing pre-processed data using a generative AI model to forecast demand, customer preferences, optimize delivery routes, and predict delivery times.

[0500] The "proposal generation means" is a means for generating optimal delivery routes, delivery times, and order lists based on the analysis results of the generative AI model.

[0501] The "visual display means" is a means for displaying generated proposals and analysis results in a format that is easy for managers and staff to understand.

[0502] "Retraining methods" are methods for incorporating new collected data into a generative AI model and retraining it to improve the model's accuracy.

[0503] An "autonomous vehicle" is a vehicle that can drive autonomously and deliver goods to a destination.

[0504] This invention is a food supply chain management system that uses autonomous vehicles, and aims to achieve efficient and effective data collection, data preprocessing, analysis using generative AI models, proposal generation, and visual display of results.

[0505] First, the server collects data. Specifically, it obtains location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, customer purchase history data, and order history data from partner stores and restaurants. It also simultaneously obtains trend data from external sources such as social media.

[0506] The server then preprocesses the collected data. This includes filling in missing values, detecting and correcting outliers, standardizing data formats, and removing duplicate data. For example, if there are missing sales data, a method is used to fill in estimated values ​​using other related data.

[0507] The preprocessed data is passed to a generative AI model. The server uses the generative AI model to perform detailed analysis of the data. Specific analysis items include demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. This makes it possible to predict demand for the following week based on past data, or predict the popularity of specific products.

[0508] Based on the analysis results, the server generates specific proposals, such as routes and delivery schedules for autonomous vehicles, next week's order lists and inventory management strategies for supermarkets and restaurants, and even suggestions for developing new menus and promotional campaigns.

[0509] The generated proposals are visually displayed on the terminal. The proposals are displayed in graph and list format on the terminal, making them easy for managers and staff to understand and implement. For example, a supermarket manager can check the order list on the terminal and make any necessary adjustments.

[0510] The server retrains the generative AI model every time new data is collected, improving the accuracy of the system, which allows it to provide timely and more optimal recommendations.

[0511] Specific examples

[0512] For example, if the server predicts that demand for ice cream will increase in a particular region during the summer, it will combine past sales data with local weather data to predict ice cream demand for the following week. As a result, a suggestion to order twice the usual amount will be displayed on the terminal. It may also analyze customer purchase history data to determine that a particular ice cream (such as chocolate) is very popular, and suggest launching a promotional campaign using this as a main product.

[0513] Example prompt sentence:

[0514] "Design an algorithm to predict ice cream demand in a specific region during the summer and suggest the optimal order quantity."

[0515] "Consider sales data from the past three months to forecast next week's sales of essential items and generate an optimal order list."

[0516] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0517] Step 1:

[0518] The server collects data, specifically location information from autonomous vehicles, delivery progress data, and warehouse temperature data, as well as sales data, inventory data, customer purchase history data, order history data, and external social media trend data from partner stores and restaurants. This allows the current status of the entire supply chain to be collected as data.

[0519] Input: location information from self-driving vehicles, delivery progress data, warehouse temperature data, sales data from stores and restaurants, inventory data, purchase history data, order history data, SNS trend data

[0520] Output: Raw data collected

[0521] Step 2:

[0522] The server performs preprocessing on the collected data, including filling in missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are missing sales data, it fills in the estimated value based on other related data. This ensures data consistency and quality.

[0523] Input: Raw data collected

[0524] Output: Preprocessed data

[0525] Step 3:

[0526] The server inputs the preprocessed data into a generative AI model for detailed analysis. Specifically, it performs demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. The generative AI model predicts future trends from past data and suggests optimal actions.

[0527] Input: Preprocessed data

[0528] Output: Analysis results

[0529] Step 4:

[0530] The server generates specific proposals based on the analysis results of the generative AI model, such as optimal delivery routes and schedules for autonomous vehicles, order lists and inventory management strategies for the following week, and proposals for the development of new menu items and sales promotion campaigns.

[0531] Input: Analysis results

[0532] Output: Specific proposals (optimal delivery routes and schedules, order lists, inventory management measures, proposals for developing new menu items, proposals for sales promotion campaigns)

[0533] Step 5:

[0534] The server visually displays the generated proposals on the terminal, where they are displayed in graph and list format, providing managers and staff with a format that is easy to understand. This allows managers and staff to intuitively understand the proposals and make any necessary adjustments or actions.

[0535] Input: Specific proposal

[0536] Output: Visual display (graphs, list of suggestions)

[0537] Step 6:

[0538] The server retrains the generative AI model every time new data is collected, improving the model's accuracy and its ability to provide better recommendations. The retrained model ensures that predictions and recommendations are always based on the most up-to-date data.

[0539] Input: New data

[0540] Output: Retrained generative AI model

[0541] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0542] This invention relates to a system that combines generative AI technology with an emotion engine that recognizes user emotions to efficiently manage food supply chains in the retail and restaurant industries. The system provides functions for data collection, preprocessing, analysis, recommendation generation, visual display, continuous learning, and emotion recognition.

[0543] Program processing flow

[0544] 1. Data Collection

[0545] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants via API. It also simultaneously collects social media and local trend data and stores it as external environment data.

[0546] 2. Data Preprocessing

[0547] The server performs various preprocessing operations on the collected data, such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, the server fills in estimated values ​​based on similar data from the past.

[0548] 3. Demand forecasting

[0549] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[0550] 4. Customer preference analysis

[0551] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[0552] 5. Trend Analysis

[0553] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[0554] 6. Emotion recognition

[0555] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to determine whether they are expressing positive or negative emotions. For example, if a customer's posts are positive, it can suggest new product information or promotions to the customer.

[0556] 7. Emotion-based suggestion generation

[0557] The server generates customized marketing messages and new product suggestions based on the customer's emotional state analyzed by the emotion engine, and if the customer is dissatisfied, suggests special offers or services to resolve the problem.

[0558] 8. Results display

[0559] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can review the display and make adjustments as necessary.

[0560] 9. Continuous learning

[0561] The server retrains the generative AI model and emotion engine as new data is collected, allowing the system to become more accurate over time and capable of providing more optimal recommendations.

[0562] Specific examples

[0563] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[0564] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[0565] Furthermore, the emotion engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the customer an invitation to a tasting event for the new menu. The suggestion is displayed on the terminal, and the restaurant manager sends the invitation message to the customer according to the suggestion.

[0566] In this way, by combining generative AI technology with emotion recognition, the system can help retail and food service businesses efficiently and effectively manage their food supply chains, increasing customer satisfaction and sales.

[0567] The processing flow will be explained below.

[0568] Program processing steps

[0569] Step 1: Data collection

[0570] The server collects various data from each retail and restaurant establishment. Specifically, it periodically obtains sales data, inventory data, customer purchase history data, and customer order history data via API. At the same time, it also collects social media and regional trend data and stores it as external environment data for analysis.

[0571] Step 2: Data Preprocessing

[0572] The server performs preprocessing on the collected data. During this process, it complements missing values ​​in the data, detects and corrects outliers, standardizes data formats, and removes duplicate data. For example, if there is a defect in sales data, it can infer the defect from other related data and complement it.

[0573] Step 3: Demand forecast

[0574] The server then inputs the pre-processed data into a generative AI model to forecast demand for the next week or month. Based on past sales data, inventory data, and regional trend data, the server makes detailed forecasts of demand for specific products and days.

[0575] Step 4: Customer preference analysis

[0576] The server analyzes customer purchase and order history data to identify which products and categories a particular customer prefers, and uses this information to generate personalized marketing messages promoting specific products and services.

[0577] Step 5: Trend analysis

[0578] The server analyzes social media data and regional trend data to identify products and categories that are likely to be popular now and in the near future. Based on this information, it makes suggestions for adding new products or running promotional campaigns, for example.

[0579] Step 6: Emotion Recognition

[0580] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to identify positive and negative emotions. For example, if a customer has positive feelings about a new product, that emotion can be used to promote the new product.

[0581] Step 7: Sentiment-based suggestion generation

[0582] The server generates customized marketing messages and new product proposals based on the customer's emotional state analyzed by the emotion engine. If a customer is dissatisfied with a particular product or service, the server will suggest special offers or services to resolve the issue.

[0583] Step 8: View the results

[0584] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can check the display and adjust and implement the proposals as necessary.

[0585] Step 9: Take Action

[0586] Users (store managers and staff) follow instructions on the terminal to actually place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and sending customized messages to specific customers based on the results of sentiment analysis.

[0587] Step 10: Continuous learning

[0588] The server retrains the generative AI model and sentiment engine based on new data collected, further improving accuracy over time, giving the system the ability to provide increasingly accurate suggestions.

[0589] These are the specific processing steps in the AI ​​Food Partner system that combines an emotion engine. This is expected to improve the efficiency of the food supply chain, improve customer satisfaction, and increase sales.

[0590] Example 2

[0591] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0592] Efficient management of food supply chains in the retail and restaurant industries requires the collection and analysis of a wide variety of data, and marketing that takes customer emotions into consideration is also required. Conventional systems have difficulty processing this data in a centralized manner, which can lead to insufficient demand forecasting and analysis of customer preferences. Furthermore, it is difficult to generate proposals that take customer emotions into consideration, leaving challenges in improving customer satisfaction and sales.

[0593] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating suggestions based on the analysis results, means for visually displaying the suggestions, means for incorporating new data and retraining the generative AI model, means for analyzing emotions using an emotion engine that recognizes the emotional state of customers, and means for generating customized suggestions based on the analyzed emotional state. This enables efficient management of the food supply chain and improves customer satisfaction and sales by providing suggestions based on customer preferences and emotions.

[0594] The "means of collecting data" refers to a mechanism for collecting sales data, inventory data, customer purchase history data, order history data, and external trend data and social media data from each retail and food service store.

[0595] The "preprocessing means" is a processing mechanism for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[0596] A "means for analyzing using a generative AI model" is a mechanism that uses preprocessed data as input and analyzes and predicts the data using a specific algorithm.

[0597] The "means for generating proposals" is a mechanism for creating customized marketing messages and product proposals based on the analysis results and sentiment analysis results.

[0598] The "visual display means" refers to a display device and its control mechanism for visually presenting the generated proposals and analysis results to managers and staff.

[0599] A "means for retraining" is a mechanism for retraining a generative AI model to improve its accuracy each time new data is acquired.

[0600] An "emotion engine" is an analytical device that analyzes a customer's emotional state from text and other data and determines whether the emotion is positive, negative, or neutral.

[0601] The "means for generating customized offers" is a mechanism for generating offers and marketing messages that are suited to individual customers based on the analyzed emotional state.

[0602] MODE FOR CARRYING OUT THE INVENTION

[0603] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries, improving customer satisfaction and sales. The system provides functions for data collection, data preprocessing, analysis using generative AI models, recommendation generation, visual display, continuous learning, emotion recognition, and customized recommendations.

[0604] Hardware and software used

[0605] server

[0606] The server is a computer that runs the following software:

[0607] Data collection software: Acquires data via API, collects data using web crawling technology, etc.

[0608] Data preprocessing software: Using data analysis tools such as Pandas and Scikit-learn, we perform tasks such as imputing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[0609] Generative AI models: Data analysis and demand forecasting, including time series forecasting models built using PyTorch and TensorFlow.

[0610] Sentiment recognition engine: Analyze customer sentiment using a sentiment analysis engine such as Google Cloud Natural Language API.

[0611] Retraining capabilities: Ability to retrain generative AI models and sentiment analysis engines with each information update.

[0612] Terminal

[0613] The terminal is a computer or tablet for running the following software:

[0614] Visual display software: Use libraries such as D3.js and Chart.js to visually display recommendations and analysis results.

[0615] Specific examples

[0616] For example, when a local supermarket needs to forecast demand for a particular product, the server collects past sales data, inventory data, and local trend data (such as past weather information and social media trend information). This data is then preprocessed and input into a generative AI model. If the result predicts that demand next week will be twice the usual amount, the device will visually display the detailed forecast along with a suggestion to double the order quantity.

[0617] Furthermore, when analyzing social media posts using emotion recognition, it can be determined that a particular dish (e.g., homemade pizza) is popular in a region. Based on this information, the server generates a proposal to add homemade pizza to the menu and run a campaign on social media. The proposal and specific campaign methods are visually displayed on the device.

[0618] Furthermore, the emotion recognition engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the relevant customer an invitation message to a tasting event for the new menu. The suggestion is displayed on the device, and the user can send the message to the customer according to the suggestion.

[0619] Prompt Sentence Examples

[0620] Demand forecast prompt statement

[0621] python

[0622] past_sales_data = get_past_sales_data()

[0623] trend_data = get_trend_data()

[0624] input_data = combine_data(past_sales_data, trend_data)

[0625] predicted_demand = ai_model.predict(input_data)

[0626] print(predicted_demand)

[0627] Customer Preference Analysis Prompt

[0628] python

[0629] customer_purchase_history = get_customer_purchase_history(customer_id)

[0630] recommendations = recommend_products(customer_purchase_history)

[0631] print(recommendations)

[0632] As described above, by combining generative AI technology with emotion recognition, this invention can help retail and food service businesses efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[0633] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0634] Program processing flow and specific explanation

[0635] Step 1: Data collection

[0636] The server collects data from each retail and restaurant establishment, including the following data:

[0637] Input: Sales data, inventory data, customer purchase history, order history, social media data, and regional trend data obtained using API and web crawling technology.

[0638] Data processing: Store data is obtained in real time through APIs, and social media and trend data is collected by web crawling.

[0639] Output: Various datasets stored in a database.

[0640] Specifically, the server retrieves supermarket sales data through an API call and executes a crawling script to collect social media trends.

[0641] Step 2: Data Preprocessing

[0642] The server performs pre-processing on the collected data.

[0643] Input: Raw data collected in step 1.

[0644] Data processing: Using Pandas and Scikit-learn, we perform missing value imputation, outlier detection and correction, data format standardization, and duplicate data removal.

[0645] Output: A preprocessed and clean dataset.

[0646] As a specific operation, for example, if there is a missing value in the sales data, the server will calculate the average value from similar past data and fill in the missing value.

[0647] Step 3: Demand forecast

[0648] The server uses a generative AI model to predict demand.

[0649] Input: The preprocessed dataset from step 2.

[0650] Data Computing: Use time series forecasting models to forecast demand for the next week. Use generative AI models in PyTorch and TensorFlow.

[0651] Output: Forecasted demand data.

[0652] Specifically, the server inputs past sales and inventory data into the generative AI model, obtaining a prediction result such as, "Demand for ice cream will be twice the usual amount next Tuesday."

[0653] Step 4: Customer preference analysis

[0654] The server analyzes the customer's purchasing history and other information to identify individual preferences.

[0655] Input: Customer purchase and order history data preprocessed in step 2.

[0656] Data computation: Using Python and Scikit-learn, data is clustered to identify customer preferences.

[0657] Output: A list of each customer's preferred products and categories.

[0658] Specifically, the server may identify, for example, that a customer has frequently purchased chocolate-related products over the past six months, and then use that information to suggest a special campaign.

[0659] Step 5: Trend analysis

[0660] The server analyzes social media data and regional trend data to identify trends.

[0661] Input: Social media data and local trend data collected in Step 1.

[0662] Data Computing: Use NLTK and Spacy to perform text analysis and identify trending products and categories.

[0663] Output: A list of trending items and categories.

[0664] Specifically, the server detects trends such as "homemade pizza is becoming a hot topic on local social media."

[0665] Step 6: Emotion Recognition

[0666] The server analyzes the customer's emotions using an emotion engine.

[0667] Input: Text data such as social media posts and reviews.

[0668] Data calculation: Uses Google Cloud Natural Language API to identify positive, negative, and neutral sentiment.

[0669] Output: Customer sentiment analysis data.

[0670] As a specific operation, the server obtains a result such as "There are many posts expressing very positive sentiment toward this new dessert menu."

[0671] Step 7: Sentiment-based suggestion generation

[0672] The server generates customized suggestions based on the emotional state analyzed by the emotion engine.

[0673] Input: Sentiment analysis result data from step 6.

[0674] Data Computing: Generate customized marketing messages and new product offers using Python and JavaScript.

[0675] Output: A customized suggestion message.

[0676] As a specific example of its operation, the server may make a suggestion such as "generating an invitation message to a tasting event of a new menu for customers who have expressed positive emotions."

[0677] Step 8: View the results

[0678] The terminal visually displays the suggestions received from the server to the manager or staff.

[0679] Input: Proposed data and analysis results received from the server.

[0680] Display: Display in graph or list format using D3.js or Chart.js.

[0681] Output: Visually displayed recommendations and analysis results.

[0682] Specifically, the device will display a message with a graph, such as, "Demand for ice cream will be twice the usual amount next Tuesday, so please double your order quantity."

[0683] Step 9: Continuous learning

[0684] The server takes in new data and retrains the generative AI model and emotion engine.

[0685] Input: The newly collected dataset.

[0686] Data computation: Update model parameters and retrain using TensorFlow or PyTorch.

[0687] Output: An improved generative AI model and emotion engine.

[0688] Specifically, as data is updated weekly, sales trends and changes in customer sentiment are reflected in the model, improving the accuracy of the next forecast.

[0689] (Application example 2)

[0690] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0691] The retail and restaurant industries require systems that efficiently manage food supply chains while taking into account customer emotions. However, current systems are unable to generate proposals that fully reflect customer emotions and preferences, posing challenges in improving customer satisfaction and making appropriate demand forecasts. Therefore, a method is needed that analyzes customers' emotional states using a variety of data and makes personalized proposals.

[0692] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0693] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating recommendations based on the analysis results, means for visually displaying the recommendations, means for incorporating new data and retraining the generative AI model, means for analyzing user sentiment, and means for generating recommendations based on the sentiment analysis results, thereby enabling personalized food supply chain management that takes into account the emotional state of customers.

[0694] "Means for collecting data" refers to a mechanism for obtaining sales data, inventory data, customer purchasing history data, social media data, and regional trend data from each retail and food service store.

[0695] The "preprocessing means" is a mechanism for complementing missing values ​​in collected data, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[0696] "Means for analysis using a generative AI model" refers to a mechanism for performing demand forecasting and customer preference analysis using a generative AI model based on preprocessed data.

[0697] The "means for generating proposals" is a mechanism for creating personalized marketing messages and new product proposals based on the analysis results and sentiment analysis results of the generative AI model.

[0698] The "visual display means" is a mechanism for visually displaying the proposals received from the server to the administrator or staff in the form of graphs, lists, etc.

[0699] A "means for retraining the generative AI model" is a mechanism for retraining the generative AI model each time new data is collected, improving its accuracy over time.

[0700] "Means for analyzing user emotions" refers to a mechanism that analyzes customer purchasing behavior and social media posting data to determine whether emotions are positive or negative.

[0701] The "means for generating proposals based on the results of sentiment analysis" is a mechanism for generating customized marketing messages and new product proposals for customers based on the results of sentiment analysis.

[0702] "Means for collecting retail sales data, inventory data, and customer purchase history data" refers to mechanisms such as APIs and sensors for obtaining sales, inventory, and customer purchase history from retail stores.

[0703] "Means for collecting food service sales data and customer order history data" refers to mechanisms such as APIs and sensors for obtaining sales and specific customer order histories from food service establishments.

[0704] "Means for collecting trend data from outside" refers to a mechanism for obtaining regional trend information from social media and the Internet.

[0705] The "means for imputing missing values ​​and detecting and correcting outliers" is a mechanism for estimating and imputing missing values, detecting outliers, and appropriately correcting them for collected data.

[0706] The "means for unifying data formats and deleting duplicate data" is a mechanism for unifying data of different formats and deleting duplicate data.

[0707] The "means for generating personalized proposals based on the results of sentiment analysis" is a mechanism for using the results of sentiment analysis to make personalized product and service proposals to individual customers.

[0708] MODE FOR CARRYING OUT THE INVENTION

[0709] This invention is a system for efficiently and effectively managing food supply chains in the retail and restaurant industries. The system combines generative AI technology with an emotion engine that recognizes user emotions to provide data collection, preprocessing, analysis, recommendation generation, visual display, and continuous learning functions. The system can also be implemented as a smartphone application. Specific embodiments of the system are described below.

[0710] Hardware and Software

[0711] Hardware

[0712] Smartphone (iOS or Android)

[0713] Software Framework

[0714] requests: for API calls

[0715] json: For working with JSON data

[0716] textblob: for sentiment analysis

[0717] sklearn: For data preprocessing and demand forecasting

[0718] pandas: for data preprocessing

[0719] Data flow

[0720] The server processes the data through the following steps:

[0721] 1. Data Collection

[0722] The server collects sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant establishments via API.

[0723] 2. Data Preprocessing

[0724] The server complements missing values ​​in the collected data, detects and corrects outliers, standardizes data formats, and deletes duplicate data.

[0725] 3. Demand forecasting

[0726] The server inputs the pre-processed data into a generative AI model to predict future demand.

[0727] 4. Emotion recognition

[0728] The server uses an emotion engine to analyze the user's emotional state and determine whether the emotion is positive or negative based on social media posts and purchase history data.

[0729] 5. Proposal generation

[0730] The server generates personalized marketing messages and new product suggestions based on the analysis results of the generative AI model and the results of sentiment analysis.

[0731] 6. Results display

[0732] The terminal visually displays the suggestions received from the server to managers and staff in the form of graphs and lists.

[0733] 7. Continuous learning

[0734] The server retrains the generative AI model and emotion engine every time new data is collected.

[0735] Specific examples

[0736] For example, suppose a user enters the text "I'm feeling really down today..." To parse this text, the server does the following:

[0737] 1. Emotion recognition

[0738] The server performs sentiment analysis using the TextBlob library and returns a negative sentiment score.

[0739] 2. Proposal generation

[0740] The server generates a suggestion, "Comfort food specially for you!" based on the negative sentiment score, past purchase history, and trend data.

[0741] The suggestions are visually displayed to the user in an app on their device, providing them with food delivery suggestions tailored to their emotional state that day.

[0742] Prompt Sentence Examples

[0743] An example of a prompt sentence to input to the generative AI model is:

[0744] "Create recommended food delivery suggestions based on the user's purchasing history and sentiment score. For example, if a user types, 'I am feeling so happy today!', provide suggestions based on positive sentiment."

[0745] Examples include:

[0746] In this way, the system of the present invention improves customer satisfaction and realizes efficient management of the food supply chain by providing personalized suggestions based on the user's emotional state.

[0747] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0748] Step 1:

[0749] Data collection

[0750] The server uses APIs to obtain sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant stores. This data is collected in JSON format. The input is the store's API key, and the output is a JSON-formatted dataset.

[0751] Step 2:

[0752] Data Preprocessing

[0753] The server performs missing value completion, outlier detection and correction, data format standardization, and duplicate data removal for the collected data. The input is the JSON dataset obtained in step 1. For each data entry, if a missing value exists, it is completed based on similar data, and outliers are corrected according to certain criteria. The output is a clean dataset.

[0754] Step 3:

[0755] Demand forecasting

[0756] The server inputs the preprocessed data into a generative AI model to predict future demand. The input is the preprocessed dataset, and the output is the demand forecast data for the next week. Specifically, it uses past sales data, inventory data, and regional trend data to predict demand for a specific day or specific product.

[0757] Step 4:

[0758] emotion recognition

[0759] The server uses an emotion engine to analyze the user's emotional state. It determines whether the emotion is positive or negative by analyzing social media posts and purchase history data. The input is social media posts and purchase history data, and the output is an emotion score.

[0760] Step 5:

[0761] Proposal generation

[0762] The server generates personalized marketing messages and new product proposals based on the analysis results of the generative AI model and the sentiment analysis results. The input is demand forecast data and sentiment scores, and the output is personalized proposals. Specifically, it prepares proposals with rewards for positive sentiment scores and includes consolation messages for negative sentiment scores.

[0763] Step 6:

[0764] Results display

[0765] The terminal visually displays the proposals received from the server in graph or list format to managers and staff. The input is the proposal data from the server, and the output is the visually displayed information. Specifically, it displays campaign proposals and customer support methods on a dashboard for managers.

[0766] Step 7:

[0767] Continuous Learning

[0768] The server retrains the generative AI model and emotion engine every time new data is collected. The input is a new dataset, and the output is a retrained model. Specifically, the model is periodically updated to improve its accuracy.

[0769] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0770] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0771] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0772] [Third embodiment]

[0773] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0774] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0775] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0776] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0777] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0778] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0779] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0780] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0781] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0782] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0783] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0784] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0785] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative AI techniques for data collection, pre-processing, analysis, recommendation generation, and visual display.

[0786] Program processing flow

[0787] 1. Data Collection

[0788] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants. It also takes into account the influence of external factors by simultaneously obtaining social media and local trend data.

[0789] 2. Data Preprocessing

[0790] The server performs various preprocessing operations on the collected data, such as filling in missing values, detecting and correcting outliers, standardizing data, and removing duplicate data. For example, if there are gaps in sales data, the server fills in the estimated value based on other related data.

[0791] 3. Data analysis

[0792] The server inputs the preprocessed data into a generative AI model for detailed analysis. This includes demand forecasting and customer preference analysis. For example, it predicts next week's demand based on past data and extracts patterns of products preferred by specific customers. It also analyzes regional trends and predicts the popularity of specific products.

[0793] 4. Proposal generation

[0794] The server generates specific recommendations based on the analysis results of the generative AI model, such as next week's ordering list and inventory management measures for a supermarket, or new menu development and promotional campaigns for a restaurant.

[0795] 5. Display results

[0796] The terminals provide managers and staff with visual suggestions, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can view next week's order list on the terminal and make any necessary adjustments.

[0797] 6. Continuous learning

[0798] The server retrains the generative AI model as new data is collected, allowing the system to become more accurate over time and provide increasingly optimal recommendations.

[0799] Specific examples

[0800] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[0801] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[0802] In this way, the system enables retailers and food service businesses to efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[0803] The processing flow will be explained below.

[0804] Program processing steps

[0805] Step 1: Data collection

[0806] The server obtains sales data, inventory data, customer purchase history data, and customer order history data from each retail and restaurant store via API. It also simultaneously collects social media and regional trend data and stores it as external environment data.

[0807] Step 2: Data Preprocessing

[0808] The server performs preprocessing on the collected data. Specifically, it performs operations such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, it fills in estimated values ​​based on similar data from the past.

[0809] Step 3: Demand forecast

[0810] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[0811] Step 4: Customer preference analysis

[0812] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[0813] Step 5: Trend analysis

[0814] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[0815] Step 6: Proposal Generation

[0816] Based on the analysis results, the server generates proposals for each store, such as the optimal product list, order quantities, and sales promotion measures. For example, it generates next week's order list and inventory management measures for a supermarket, and new menu items and sales promotion campaigns for a restaurant.

[0817] Step 7: View the results

[0818] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, and campaign proposals. Managers can review the display and make adjustments as necessary.

[0819] Step 8: Take action

[0820] Users (store managers and staff) follow instructions on the terminal to place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and planning and implementing sales promotion campaigns.

[0821] Step 9: Continuous learning

[0822] The server retrains the generative AI model based on new data collected, thereby maintaining the accuracy of the predictive algorithm and improving the system's performance over time.

[0823] These are the specific processing steps in the AI ​​Food Partner program, which will optimize the food supply chain, improve efficiency in the retail and restaurant industries, and increase customer satisfaction.

[0824] Example 1

[0825] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0826] Managing food supply chains in the retail and restaurant industries requires efficient collection and analysis of a wide range of data and the generation of optimal proposals. However, conventional methods require a great deal of time and effort for data collection and preprocessing, and the accuracy of the analysis is limited. As a result, proper inventory management and demand forecasting are difficult, resulting in excess inventory and lost opportunities. The present invention aims to provide a system that solves these problems.

[0827] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0828] In this invention, the server includes a means for collecting data, a means for preprocessing the collected data, and a means for analyzing the preprocessed data with a generative artificial intelligence model. This allows for automatic collection of data from multiple data sources and highly accurate preprocessing and analysis, thereby enabling efficient inventory management and demand forecasting.

[0829] "Means for collecting data" refers to communication modules and software for automatically acquiring various data, including sales data, inventory data, customer purchase history data, and order history data, from each retail and food service store.

[0830] The "means for preprocessing the collected data" refers to algorithms or software modules for completing missing values ​​in the collected data, detecting and correcting outliers, standardizing data formats, and removing duplicate data.

[0831] "Means of analysis using generative artificial intelligence models" refers to machine learning models and algorithms that use preprocessed data to perform detailed analysis such as demand forecasting and customer preference analysis.

[0832] The "means for generating proposals" refers to algorithms or software modules that create specific action plans, such as inventory management measures, new menu proposals, and sales promotion campaigns, based on the analysis results of the generative artificial intelligence model.

[0833] "Means for visually displaying proposals" refers to a dashboard or user interface that displays the generated proposals and analysis results in graphs or tables in a way that is easy for managers and staff to understand.

[0834] "Means for incorporating new data and retraining the generated AI model" refers to algorithms or software that use new data collected periodically to retrain the AI ​​model to improve its accuracy.

[0835] "Retail sales data" refers to data that includes transaction information when a product is sold at a retail store.

[0836] "Inventory data" refers to data that includes information indicating the inventory status of a retail store or restaurant at a specific point in time.

[0837] "Customer purchase history data" is data that includes records of products and services purchased by individual customers in the past.

[0838] "Customer order history data" refers to data in the food and beverage industry that includes records of food and drink orders that customers have made in the past.

[0839] "Trend data" is external data that shows consumption trends and fashions in a specific region or period, and is obtained from social media and local information services.

[0840] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative artificial intelligence techniques to collect, preprocess, analyze, generate recommendations, and visually display data. The details of this system are described below.

[0841] The server uses APIs and communication modules to collect data from each retail and restaurant chain. For example, it obtains sales and inventory data from a supermarket's sales database, and customer order history from a restaurant's ordering system. It also collects social media data and local trend data in parallel, taking into account the influence of external factors.

[0842] The collected data is preprocessed by the server. This preprocessing includes filling in missing data, correcting inventory data, standardizing data formats, and removing duplicate data. Specifically, if there are missing values ​​in sales data, they are filled in based on other related data. Anomaly detection is performed using TensorFlow and PyTorch algorithms to maintain data quality.

[0843] The server then inputs the preprocessed data into generative artificial intelligence models to perform demand forecasting and customer preference analysis. This analysis uses natural language processing (NLP) models and machine learning algorithms. For example, a model is run to predict next week's demand based on past sales data and local weather data. Another model is run to extract customer purchasing patterns and predict the popularity of certain products.

[0844] Based on the analysis results, the server generates specific proposals, such as next week's order list and inventory management for supermarkets, and new menu development and promotion campaigns for restaurants, to ensure smooth supply chain management and maximize sales opportunities.

[0845] The terminals display visual suggestions to managers and staff, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can see next week's order list displayed on the terminal and make necessary adjustments accordingly.

[0846] The system improves its accuracy through continuous learning: as new data is collected, the server retrains the generative artificial intelligence model to provide more optimal recommendations, allowing the system to evolve over time and become even more effective.

[0847] Examples of specific prompts for generative AI models

[0848] The following sentence can be used as an example of a prompt sentence to be input to the generative artificial intelligence model:

[0849] Example prompt 1:

[0850] "Predict next week's demand for ice cream based on weekly sales data from the past year and local weather data."

[0851] Example prompt 2:

[0852] "Using customer purchase history and social media trend data from the past three months, please suggest new menu items that will be most popular next month."

[0853] This makes it possible to automatically generate specific and effective proposals for the retail and food service industries.

[0854] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0855] Step 1: Data collection

[0856] The server collects various data from each retail and restaurant chain. Specifically, the server accesses the supermarket's sales database via API to obtain sales and inventory data. It also obtains customer order history from the restaurant's ordering system, and uses social media APIs and local information service APIs to collect social media data and local trend data. The input is the various data received from the API, and the output is raw collected data that is stored in a central database within the server.

[0857] Step 2: Data Preprocessing

[0858] The server preprocesses the collected data. First, if there is missing data, it fills it in based on other relevant data. Next, it uses statistical methods and machine learning algorithms to detect and correct outliers. If the data format is different, it is converted into a consistent format, and if there is duplicate data, it is removed. The input is the raw data collected in step 1, and the output is the clean data after preprocessing.

[0859] Step 3: Data analysis

[0860] The server inputs the preprocessed data into a generative artificial intelligence model for further analysis. For example, it uses TensorFlow and PyTorch to run demand forecasting and customer preference analysis models. Specifically, it uses past sales data and weather data to predict future demand and analyze customer purchasing patterns. The input is the preprocessed data, and the output is the analysis results, which include predicted sales volume, a list of popular products, and trend analysis results.

[0861] Step 4: Proposal Generation

[0862] The server generates specific proposals based on the analysis results generated. For example, it provides a supermarket with an order list for the next week based on the results of demand forecasting, or creates a promotional campaign for a specific customer group based on the results of customer preference analysis. It also proposes new menu additions to restaurants based on regional trends. The input is the analysis results generated in step 3, and the output is specific proposals.

[0863] Step 5: View the results

[0864] The terminal visualizes and displays the proposals provided by the server. It displays sales forecasts in graph form, order lists in list form, and campaign proposals on a dashboard accessible to managers and staff. For example, a supermarket manager can check next week's order list on the terminal and make necessary adjustments. The input is the proposal generated in step 4, and the output is the visual representation of the proposal.

[0865] Step 6: Continuous learning

[0866] The server retrains the generative AI model using newly collected data, which improves the model's accuracy over time and enables it to provide more accurate and useful suggestions. The input is the latest newly collected data, and the output is a retrained, highly accurate generative AI model.

[0867] (Application example 1)

[0868] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0869] Traditional food supply chain management systems are limited to collecting and analyzing data within stores and warehouses, and do not optimize the delivery process. This makes it difficult to select efficient routes during delivery or monitor progress in real time, which can lead to placement losses and deterioration of storage quality. Furthermore, delays in demand forecasting and response, as well as difficulties in optimal ordering and inventory management, lead to lost sales opportunities and excess inventory.

[0870] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0871] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating proposals based on the analysis results, means for visually displaying the proposals, means for incorporating new data and retraining the generative AI model, and means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles. This enables real-time data collection and analysis during the delivery process, making it possible to propose optimal delivery routes and times, and to realize efficient ordering and inventory management based on demand forecasts.

[0872] "Data collection means" refers to means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, and customer purchase history data from retail and food service businesses.

[0873] The "preprocessing means" refers to means for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[0874] A "generative AI model analysis means" is a means for analyzing pre-processed data using a generative AI model to forecast demand, customer preferences, optimize delivery routes, and predict delivery times.

[0875] The "proposal generation means" is a means for generating optimal delivery routes, delivery times, and order lists based on the analysis results of the generative AI model.

[0876] The "visual display means" is a means for displaying generated proposals and analysis results in a format that is easy for managers and staff to understand.

[0877] "Retraining methods" are methods for incorporating new collected data into a generative AI model and retraining it to improve the model's accuracy.

[0878] An "autonomous vehicle" is a vehicle that can drive autonomously and deliver goods to a destination.

[0879] This invention is a food supply chain management system that uses autonomous vehicles, and aims to achieve efficient and effective data collection, data preprocessing, analysis using generative AI models, proposal generation, and visual display of results.

[0880] First, the server collects data. Specifically, it obtains location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, customer purchase history data, and order history data from partner stores and restaurants. It also simultaneously obtains trend data from external sources such as social media.

[0881] The server then preprocesses the collected data. This includes filling in missing values, detecting and correcting outliers, standardizing data formats, and removing duplicate data. For example, if there are missing sales data, a method is used to fill in estimated values ​​using other related data.

[0882] The preprocessed data is passed to a generative AI model. The server uses the generative AI model to perform detailed analysis of the data. Specific analysis items include demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. This makes it possible to predict demand for the following week based on past data, or predict the popularity of specific products.

[0883] Based on the analysis results, the server generates specific proposals, such as routes and delivery schedules for autonomous vehicles, next week's order lists and inventory management strategies for supermarkets and restaurants, and even suggestions for developing new menus and promotional campaigns.

[0884] The generated proposals are visually displayed on the terminal. The proposals are displayed in graph and list format on the terminal, making them easy for managers and staff to understand and implement. For example, a supermarket manager can check the order list on the terminal and make any necessary adjustments.

[0885] The server retrains the generative AI model every time new data is collected, improving the accuracy of the system, which allows it to provide timely and more optimal recommendations.

[0886] Specific examples

[0887] For example, if the server predicts that demand for ice cream will increase in a particular region during the summer, it will combine past sales data with local weather data to predict ice cream demand for the following week. As a result, a suggestion to order twice the usual amount will be displayed on the terminal. It may also analyze customer purchase history data to determine that a particular ice cream (such as chocolate) is very popular, and suggest launching a promotional campaign using this as a main product.

[0888] Example prompt sentence:

[0889] "Design an algorithm to predict ice cream demand in a specific region during the summer and suggest the optimal order quantity."

[0890] "Consider sales data from the past three months to forecast next week's sales of essential items and generate an optimal order list."

[0891] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0892] Step 1:

[0893] The server collects data, specifically location information from autonomous vehicles, delivery progress data, and warehouse temperature data, as well as sales data, inventory data, customer purchase history data, order history data, and external social media trend data from partner stores and restaurants. This allows the current status of the entire supply chain to be collected as data.

[0894] Input: location information from self-driving vehicles, delivery progress data, warehouse temperature data, sales data from stores and restaurants, inventory data, purchase history data, order history data, SNS trend data

[0895] Output: Raw data collected

[0896] Step 2:

[0897] The server performs preprocessing on the collected data, including filling in missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are missing sales data, it fills in the estimated value based on other related data. This ensures data consistency and quality.

[0898] Input: Raw data collected

[0899] Output: Preprocessed data

[0900] Step 3:

[0901] The server inputs the preprocessed data into a generative AI model for detailed analysis. Specifically, it performs demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. The generative AI model predicts future trends from past data and suggests optimal actions.

[0902] Input: Preprocessed data

[0903] Output: Analysis results

[0904] Step 4:

[0905] The server generates specific proposals based on the analysis results of the generative AI model, such as optimal delivery routes and schedules for autonomous vehicles, order lists and inventory management strategies for the following week, and proposals for the development of new menu items and sales promotion campaigns.

[0906] Input: Analysis results

[0907] Output: Specific proposals (optimal delivery routes and schedules, order lists, inventory management measures, proposals for developing new menu items, proposals for sales promotion campaigns)

[0908] Step 5:

[0909] The server visually displays the generated proposals on the terminal, where they are displayed in graph and list format, providing managers and staff with a format that is easy to understand. This allows managers and staff to intuitively understand the proposals and make any necessary adjustments or actions.

[0910] Input: Specific proposal

[0911] Output: Visual display (graphs, list of suggestions)

[0912] Step 6:

[0913] The server retrains the generative AI model every time new data is collected, improving the model's accuracy and its ability to provide better recommendations. The retrained model ensures that predictions and recommendations are always based on the most up-to-date data.

[0914] Input: New data

[0915] Output: Retrained generative AI model

[0916] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0917] This invention relates to a system that combines generative AI technology with an emotion engine that recognizes user emotions to efficiently manage food supply chains in the retail and restaurant industries. The system provides functions for data collection, preprocessing, analysis, recommendation generation, visual display, continuous learning, and emotion recognition.

[0918] Program processing flow

[0919] 1. Data Collection

[0920] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants via API. It also simultaneously collects social media and local trend data and stores it as external environment data.

[0921] 2. Data Preprocessing

[0922] The server performs various preprocessing operations on the collected data, such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, the server fills in estimated values ​​based on similar data from the past.

[0923] 3. Demand forecasting

[0924] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[0925] 4. Customer preference analysis

[0926] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[0927] 5. Trend Analysis

[0928] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[0929] 6. Emotion recognition

[0930] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to determine whether they are expressing positive or negative emotions. For example, if a customer's posts are positive, it can suggest new product information or promotions to the customer.

[0931] 7. Emotion-based suggestion generation

[0932] The server generates customized marketing messages and new product suggestions based on the customer's emotional state analyzed by the emotion engine, and if the customer is dissatisfied, suggests special offers or services to resolve the problem.

[0933] 8. Results display

[0934] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can review the display and make adjustments as necessary.

[0935] 9. Continuous learning

[0936] The server retrains the generative AI model and emotion engine as new data is collected, allowing the system to become more accurate over time and capable of providing more optimal recommendations.

[0937] Specific examples

[0938] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[0939] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[0940] Furthermore, the emotion engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the customer an invitation to a tasting event for the new menu. The suggestion is displayed on the terminal, and the restaurant manager sends the invitation message to the customer according to the suggestion.

[0941] In this way, by combining generative AI technology with emotion recognition, the system can help retail and food service businesses efficiently and effectively manage their food supply chains, increasing customer satisfaction and sales.

[0942] The processing flow will be explained below.

[0943] Program processing steps

[0944] Step 1: Data collection

[0945] The server collects various data from each retail and restaurant establishment. Specifically, it periodically obtains sales data, inventory data, customer purchase history data, and customer order history data via API. At the same time, it also collects social media and regional trend data and stores it as external environment data for analysis.

[0946] Step 2: Data Preprocessing

[0947] The server performs preprocessing on the collected data. During this process, it complements missing values ​​in the data, detects and corrects outliers, standardizes data formats, and removes duplicate data. For example, if there is a defect in sales data, it can infer the defect from other related data and complement it.

[0948] Step 3: Demand forecast

[0949] The server then inputs the pre-processed data into a generative AI model to forecast demand for the next week or month. Based on past sales data, inventory data, and regional trend data, the server makes detailed forecasts of demand for specific products and days.

[0950] Step 4: Customer preference analysis

[0951] The server analyzes customer purchase and order history data to identify which products and categories a particular customer prefers, and uses this information to generate personalized marketing messages promoting specific products and services.

[0952] Step 5: Trend analysis

[0953] The server analyzes social media data and regional trend data to identify products and categories that are likely to be popular now and in the near future. Based on this information, it makes suggestions for adding new products or running promotional campaigns, for example.

[0954] Step 6: Emotion Recognition

[0955] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to identify positive and negative emotions. For example, if a customer has positive feelings about a new product, that emotion can be used to promote the new product.

[0956] Step 7: Sentiment-based suggestion generation

[0957] The server generates customized marketing messages and new product proposals based on the customer's emotional state analyzed by the emotion engine. If a customer is dissatisfied with a particular product or service, the server will suggest special offers or services to resolve the issue.

[0958] Step 8: View the results

[0959] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can check the display and adjust and implement the proposals as necessary.

[0960] Step 9: Take Action

[0961] Users (store managers and staff) follow instructions on the terminal to actually place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and sending customized messages to specific customers based on the results of sentiment analysis.

[0962] Step 10: Continuous learning

[0963] The server retrains the generative AI model and sentiment engine based on new data collected, further improving accuracy over time, giving the system the ability to provide increasingly accurate suggestions.

[0964] These are the specific processing steps in the AI ​​Food Partner system that combines an emotion engine. This is expected to improve the efficiency of the food supply chain, improve customer satisfaction, and increase sales.

[0965] Example 2

[0966] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0967] Efficient management of food supply chains in the retail and restaurant industries requires the collection and analysis of a wide variety of data, and marketing that takes customer emotions into consideration is also required. Conventional systems have difficulty processing this data in a centralized manner, which can lead to insufficient demand forecasting and analysis of customer preferences. Furthermore, it is difficult to generate proposals that take customer emotions into consideration, leaving challenges in improving customer satisfaction and sales.

[0968] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating suggestions based on the analysis results, means for visually displaying the suggestions, means for incorporating new data and retraining the generative AI model, means for analyzing emotions using an emotion engine that recognizes the emotional state of customers, and means for generating customized suggestions based on the analyzed emotional state. This enables efficient management of the food supply chain and improves customer satisfaction and sales by providing suggestions based on customer preferences and emotions.

[0969] The "means of collecting data" refers to a mechanism for collecting sales data, inventory data, customer purchase history data, order history data, and external trend data and social media data from each retail and food service store.

[0970] The "preprocessing means" is a processing mechanism for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[0971] A "means for analyzing using a generative AI model" is a mechanism that uses preprocessed data as input and analyzes and predicts the data using a specific algorithm.

[0972] The "means for generating proposals" is a mechanism for creating customized marketing messages and product proposals based on the analysis results and sentiment analysis results.

[0973] The "visual display means" refers to a display device and its control mechanism for visually presenting the generated proposals and analysis results to managers and staff.

[0974] A "means for retraining" is a mechanism for retraining a generative AI model to improve its accuracy each time new data is acquired.

[0975] An "emotion engine" is an analytical device that analyzes a customer's emotional state from text and other data and determines whether the emotion is positive, negative, or neutral.

[0976] The "means for generating customized offers" is a mechanism for generating offers and marketing messages that are suited to individual customers based on the analyzed emotional state.

[0977] MODE FOR CARRYING OUT THE INVENTION

[0978] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries, improving customer satisfaction and sales. The system provides functions for data collection, data preprocessing, analysis using generative AI models, recommendation generation, visual display, continuous learning, emotion recognition, and customized recommendations.

[0979] Hardware and software used

[0980] server

[0981] The server is a computer that runs the following software:

[0982] Data collection software: Acquires data via API, collects data using web crawling technology, etc.

[0983] Data preprocessing software: Using data analysis tools such as Pandas and Scikit-learn, we perform tasks such as imputing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[0984] Generative AI models: Data analysis and demand forecasting, including time series forecasting models built using PyTorch and TensorFlow.

[0985] Sentiment recognition engine: Analyze customer sentiment using a sentiment analysis engine such as Google Cloud Natural Language API.

[0986] Retraining capabilities: Ability to retrain generative AI models and sentiment analysis engines with each information update.

[0987] Terminal

[0988] The terminal is a computer or tablet for running the following software:

[0989] Visual display software: Use libraries such as D3.js and Chart.js to visually display recommendations and analysis results.

[0990] Specific examples

[0991] For example, when a local supermarket needs to forecast demand for a particular product, the server collects past sales data, inventory data, and local trend data (such as past weather information and social media trend information). This data is then preprocessed and input into a generative AI model. If the result predicts that demand next week will be twice the usual amount, the device will visually display the detailed forecast along with a suggestion to double the order quantity.

[0992] Furthermore, when analyzing social media posts using emotion recognition, it can be determined that a particular dish (e.g., homemade pizza) is popular in a region. Based on this information, the server generates a proposal to add homemade pizza to the menu and run a campaign on social media. The proposal and specific campaign methods are visually displayed on the device.

[0993] Furthermore, the emotion recognition engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the relevant customer an invitation message to a tasting event for the new menu. The suggestion is displayed on the device, and the user can send the message to the customer according to the suggestion.

[0994] Prompt Sentence Examples

[0995] Demand forecast prompt statement

[0996] python

[0997] past_sales_data = get_past_sales_data()

[0998] trend_data = get_trend_data()

[0999] input_data = combine_data(past_sales_data, trend_data)

[1000] predicted_demand = ai_model.predict(input_data)

[1001] print(predicted_demand)

[1002] Customer Preference Analysis Prompt

[1003] python

[1004] customer_purchase_history = get_customer_purchase_history(customer_id)

[1005] recommendations = recommend_products(customer_purchase_history)

[1006] print(recommendations)

[1007] As described above, by combining generative AI technology with emotion recognition, this invention can help retail and food service businesses efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[1008] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1009] Program processing flow and specific explanation

[1010] Step 1: Data collection

[1011] The server collects data from each retail and restaurant establishment, including the following data:

[1012] Input: Sales data, inventory data, customer purchase history, order history, social media data, and regional trend data obtained using API and web crawling technology.

[1013] Data processing: Store data is obtained in real time through APIs, and social media and trend data is collected by web crawling.

[1014] Output: Various datasets stored in a database.

[1015] Specifically, the server retrieves supermarket sales data through an API call and executes a crawling script to collect social media trends.

[1016] Step 2: Data Preprocessing

[1017] The server performs pre-processing on the collected data.

[1018] Input: Raw data collected in step 1.

[1019] Data processing: Using Pandas and Scikit-learn, we perform missing value imputation, outlier detection and correction, data format standardization, and duplicate data removal.

[1020] Output: A preprocessed and clean dataset.

[1021] As a specific operation, for example, if there is a missing value in the sales data, the server will calculate the average value from similar past data and fill in the missing value.

[1022] Step 3: Demand forecast

[1023] The server uses a generative AI model to predict demand.

[1024] Input: The preprocessed dataset from step 2.

[1025] Data Computing: Use time series forecasting models to forecast demand for the next week. Use generative AI models in PyTorch and TensorFlow.

[1026] Output: Forecasted demand data.

[1027] Specifically, the server inputs past sales and inventory data into the generative AI model, obtaining a prediction result such as, "Demand for ice cream will be twice the usual amount next Tuesday."

[1028] Step 4: Customer preference analysis

[1029] The server analyzes the customer's purchasing history and other information to identify individual preferences.

[1030] Input: Customer purchase and order history data preprocessed in step 2.

[1031] Data computation: Using Python and Scikit-learn, data is clustered to identify customer preferences.

[1032] Output: A list of each customer's preferred products and categories.

[1033] Specifically, the server may identify, for example, that a customer has frequently purchased chocolate-related products over the past six months, and then use that information to suggest a special campaign.

[1034] Step 5: Trend analysis

[1035] The server analyzes social media data and regional trend data to identify trends.

[1036] Input: Social media data and local trend data collected in Step 1.

[1037] Data Computing: Use NLTK and Spacy to perform text analysis and identify trending products and categories.

[1038] Output: A list of trending items and categories.

[1039] Specifically, the server detects trends such as "homemade pizza is becoming a hot topic on local social media."

[1040] Step 6: Emotion Recognition

[1041] The server analyzes the customer's emotions using an emotion engine.

[1042] Input: Text data such as social media posts and reviews.

[1043] Data calculation: Uses Google Cloud Natural Language API to identify positive, negative, and neutral sentiment.

[1044] Output: Customer sentiment analysis data.

[1045] As a specific operation, the server obtains a result such as "There are many posts expressing very positive sentiment toward this new dessert menu."

[1046] Step 7: Sentiment-based suggestion generation

[1047] The server generates customized suggestions based on the emotional state analyzed by the emotion engine.

[1048] Input: Sentiment analysis result data from step 6.

[1049] Data Computing: Generate customized marketing messages and new product offers using Python and JavaScript.

[1050] Output: A customized suggestion message.

[1051] As a specific example of its operation, the server may make a suggestion such as "generating an invitation message to a tasting event of a new menu for customers who have expressed positive emotions."

[1052] Step 8: View the results

[1053] The terminal visually displays the suggestions received from the server to the manager or staff.

[1054] Input: Proposed data and analysis results received from the server.

[1055] Display: Display in graph or list format using D3.js or Chart.js.

[1056] Output: Visually displayed recommendations and analysis results.

[1057] Specifically, the device will display a message with a graph, such as, "Demand for ice cream will be twice the usual amount next Tuesday, so please double your order quantity."

[1058] Step 9: Continuous learning

[1059] The server takes in new data and retrains the generative AI model and emotion engine.

[1060] Input: The newly collected dataset.

[1061] Data computation: Update model parameters and retrain using TensorFlow or PyTorch.

[1062] Output: An improved generative AI model and emotion engine.

[1063] Specifically, as data is updated weekly, sales trends and changes in customer sentiment are reflected in the model, improving the accuracy of the next forecast.

[1064] (Application example 2)

[1065] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1066] The retail and restaurant industries require systems that efficiently manage food supply chains while taking into account customer emotions. However, current systems are unable to generate proposals that fully reflect customer emotions and preferences, posing challenges in improving customer satisfaction and making appropriate demand forecasts. Therefore, a method is needed that analyzes customers' emotional states using a variety of data and makes personalized proposals.

[1067] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1068] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating recommendations based on the analysis results, means for visually displaying the recommendations, means for incorporating new data and retraining the generative AI model, means for analyzing user sentiment, and means for generating recommendations based on the sentiment analysis results, thereby enabling personalized food supply chain management that takes into account the emotional state of customers.

[1069] "Means for collecting data" refers to a mechanism for obtaining sales data, inventory data, customer purchasing history data, social media data, and regional trend data from each retail and food service store.

[1070] The "preprocessing means" is a mechanism for complementing missing values ​​in collected data, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[1071] "Means for analysis using a generative AI model" refers to a mechanism for performing demand forecasting and customer preference analysis using a generative AI model based on preprocessed data.

[1072] The "means for generating proposals" is a mechanism for creating personalized marketing messages and new product proposals based on the analysis results and sentiment analysis results of the generative AI model.

[1073] The "visual display means" is a mechanism for visually displaying the proposals received from the server to the administrator or staff in the form of graphs, lists, etc.

[1074] A "means for retraining the generative AI model" is a mechanism for retraining the generative AI model each time new data is collected, improving its accuracy over time.

[1075] "Means for analyzing user emotions" refers to a mechanism that analyzes customer purchasing behavior and social media posting data to determine whether emotions are positive or negative.

[1076] The "means for generating proposals based on the results of sentiment analysis" is a mechanism for generating customized marketing messages and new product proposals for customers based on the results of sentiment analysis.

[1077] "Means for collecting retail sales data, inventory data, and customer purchase history data" refers to mechanisms such as APIs and sensors for obtaining sales, inventory, and customer purchase history from retail stores.

[1078] "Means for collecting food service sales data and customer order history data" refers to mechanisms such as APIs and sensors for obtaining sales and specific customer order histories from food service establishments.

[1079] "Means for collecting trend data from outside" refers to a mechanism for obtaining regional trend information from social media and the Internet.

[1080] The "means for imputing missing values ​​and detecting and correcting outliers" is a mechanism for estimating and imputing missing values, detecting outliers, and appropriately correcting them for collected data.

[1081] The "means for unifying data formats and deleting duplicate data" is a mechanism for unifying data of different formats and deleting duplicate data.

[1082] The "means for generating personalized proposals based on the results of sentiment analysis" is a mechanism for using the results of sentiment analysis to make personalized product and service proposals to individual customers.

[1083] MODE FOR CARRYING OUT THE INVENTION

[1084] This invention is a system for efficiently and effectively managing food supply chains in the retail and restaurant industries. The system combines generative AI technology with an emotion engine that recognizes user emotions to provide data collection, preprocessing, analysis, recommendation generation, visual display, and continuous learning functions. The system can also be implemented as a smartphone application. Specific embodiments of the system are described below.

[1085] Hardware and Software

[1086] Hardware

[1087] Smartphone (iOS or Android)

[1088] Software Framework

[1089] requests: for API calls

[1090] json: For working with JSON data

[1091] textblob: for sentiment analysis

[1092] sklearn: For data preprocessing and demand forecasting

[1093] pandas: for data preprocessing

[1094] Data flow

[1095] The server processes the data through the following steps:

[1096] 1. Data Collection

[1097] The server collects sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant establishments via API.

[1098] 2. Data Preprocessing

[1099] The server complements missing values ​​in the collected data, detects and corrects outliers, standardizes data formats, and deletes duplicate data.

[1100] 3. Demand forecasting

[1101] The server inputs the pre-processed data into a generative AI model to predict future demand.

[1102] 4. Emotion recognition

[1103] The server uses an emotion engine to analyze the user's emotional state and determine whether the emotion is positive or negative based on social media posts and purchase history data.

[1104] 5. Proposal generation

[1105] The server generates personalized marketing messages and new product suggestions based on the analysis results of the generative AI model and the results of sentiment analysis.

[1106] 6. Results display

[1107] The terminal visually displays the suggestions received from the server to managers and staff in the form of graphs and lists.

[1108] 7. Continuous learning

[1109] The server retrains the generative AI model and emotion engine every time new data is collected.

[1110] Specific examples

[1111] For example, suppose a user enters the text "I'm feeling really down today..." To parse this text, the server does the following:

[1112] 1. Emotion recognition

[1113] The server performs sentiment analysis using the TextBlob library and returns a negative sentiment score.

[1114] 2. Proposal generation

[1115] The server generates a suggestion, "Comfort food specially for you!" based on the negative sentiment score, past purchase history, and trend data.

[1116] The suggestions are visually displayed to the user in an app on their device, providing them with food delivery suggestions tailored to their emotional state that day.

[1117] Prompt Sentence Examples

[1118] An example of a prompt sentence to input to the generative AI model is:

[1119] "Create recommended food delivery suggestions based on the user's purchasing history and sentiment score. For example, if a user types, 'I am feeling so happy today!', provide suggestions based on positive sentiment."

[1120] Examples include:

[1121] In this way, the system of the present invention improves customer satisfaction and realizes efficient management of the food supply chain by providing personalized suggestions based on the user's emotional state.

[1122] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1123] Step 1:

[1124] Data collection

[1125] The server uses APIs to obtain sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant stores. This data is collected in JSON format. The input is the store's API key, and the output is a JSON-formatted dataset.

[1126] Step 2:

[1127] Data Preprocessing

[1128] The server performs missing value completion, outlier detection and correction, data format standardization, and duplicate data removal for the collected data. The input is the JSON dataset obtained in step 1. For each data entry, if a missing value exists, it is completed based on similar data, and outliers are corrected according to certain criteria. The output is a clean dataset.

[1129] Step 3:

[1130] Demand forecasting

[1131] The server inputs the preprocessed data into a generative AI model to predict future demand. The input is the preprocessed dataset, and the output is the demand forecast data for the next week. Specifically, it uses past sales data, inventory data, and regional trend data to predict demand for a specific day or specific product.

[1132] Step 4:

[1133] emotion recognition

[1134] The server uses an emotion engine to analyze the user's emotional state. It determines whether the emotion is positive or negative by analyzing social media posts and purchase history data. The input is social media posts and purchase history data, and the output is an emotion score.

[1135] Step 5:

[1136] Proposal generation

[1137] The server generates personalized marketing messages and new product proposals based on the analysis results of the generative AI model and the sentiment analysis results. The input is demand forecast data and sentiment scores, and the output is personalized proposals. Specifically, it prepares proposals with rewards for positive sentiment scores and includes consolation messages for negative sentiment scores.

[1138] Step 6:

[1139] Results display

[1140] The terminal visually displays the proposals received from the server in graph or list format to managers and staff. The input is the proposal data from the server, and the output is the visually displayed information. Specifically, it displays campaign proposals and customer support methods on a dashboard for managers.

[1141] Step 7:

[1142] Continuous Learning

[1143] The server retrains the generative AI model and emotion engine every time new data is collected. The input is a new dataset, and the output is a retrained model. Specifically, the model is periodically updated to improve its accuracy.

[1144] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1145] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1146] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1147] [Fourth embodiment]

[1148] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1149] 7, a 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.

[1150] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1151] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1152] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1153] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1154] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1155] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1156] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1157] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[1158] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1159] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1160] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1161] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative AI techniques for data collection, pre-processing, analysis, recommendation generation, and visual display.

[1162] Program processing flow

[1163] 1. Data Collection

[1164] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants. It also takes into account the influence of external factors by simultaneously obtaining social media and local trend data.

[1165] 2. Data Preprocessing

[1166] The server performs various preprocessing operations on the collected data, such as filling in missing values, detecting and correcting outliers, standardizing data, and removing duplicate data. For example, if there are gaps in sales data, the server fills in the estimated value based on other related data.

[1167] 3. Data analysis

[1168] The server inputs the preprocessed data into a generative AI model for detailed analysis. This includes demand forecasting and customer preference analysis. For example, it predicts next week's demand based on past data and extracts patterns of products preferred by specific customers. It also analyzes regional trends and predicts the popularity of specific products.

[1169] 4. Proposal generation

[1170] The server generates specific recommendations based on the analysis results of the generative AI model, such as next week's ordering list and inventory management measures for a supermarket, or new menu development and promotional campaigns for a restaurant.

[1171] 5. Display results

[1172] The terminals provide managers and staff with visual suggestions, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can view next week's order list on the terminal and make any necessary adjustments.

[1173] 6. Continuous learning

[1174] The server retrains the generative AI model as new data is collected, allowing the system to become more accurate over time and provide increasingly optimal recommendations.

[1175] Specific examples

[1176] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[1177] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[1178] In this way, the system enables retailers and food service businesses to efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[1179] The processing flow will be explained below.

[1180] Program processing steps

[1181] Step 1: Data collection

[1182] The server obtains sales data, inventory data, customer purchase history data, and customer order history data from each retail and restaurant store via API. It also simultaneously collects social media and regional trend data and stores it as external environment data.

[1183] Step 2: Data Preprocessing

[1184] The server performs preprocessing on the collected data. Specifically, it performs operations such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, it fills in estimated values ​​based on similar data from the past.

[1185] Step 3: Demand forecast

[1186] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[1187] Step 4: Customer preference analysis

[1188] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[1189] Step 5: Trend analysis

[1190] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[1191] Step 6: Proposal Generation

[1192] Based on the analysis results, the server generates proposals for each store, such as the optimal product list, order quantities, and sales promotion measures. For example, it generates next week's order list and inventory management measures for a supermarket, and new menu items and sales promotion campaigns for a restaurant.

[1193] Step 7: View the results

[1194] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, and campaign proposals. Managers can review the display and make adjustments as necessary.

[1195] Step 8: Take action

[1196] Users (store managers and staff) follow instructions on the terminal to place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and planning and implementing sales promotion campaigns.

[1197] Step 9: Continuous learning

[1198] The server retrains the generative AI model based on new data collected, thereby maintaining the accuracy of the predictive algorithm and improving the system's performance over time.

[1199] These are the specific processing steps in the AI ​​Food Partner program, which will optimize the food supply chain, improve efficiency in the retail and restaurant industries, and increase customer satisfaction.

[1200] Example 1

[1201] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1202] Managing food supply chains in the retail and restaurant industries requires efficient collection and analysis of a wide range of data and the generation of optimal proposals. However, conventional methods require a great deal of time and effort for data collection and preprocessing, and the accuracy of the analysis is limited. As a result, proper inventory management and demand forecasting are difficult, resulting in excess inventory and lost opportunities. The present invention aims to provide a system that solves these problems.

[1203] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1204] In this invention, the server includes a means for collecting data, a means for preprocessing the collected data, and a means for analyzing the preprocessed data with a generative artificial intelligence model. This allows for automatic collection of data from multiple data sources and highly accurate preprocessing and analysis, thereby enabling efficient inventory management and demand forecasting.

[1205] "Means for collecting data" refers to communication modules and software for automatically acquiring various data, including sales data, inventory data, customer purchase history data, and order history data, from each retail and food service store.

[1206] The "means for preprocessing the collected data" refers to algorithms or software modules for completing missing values ​​in the collected data, detecting and correcting outliers, standardizing data formats, and removing duplicate data.

[1207] "Means of analysis using generative artificial intelligence models" refers to machine learning models and algorithms that use preprocessed data to perform detailed analysis such as demand forecasting and customer preference analysis.

[1208] The "means for generating proposals" refers to algorithms or software modules that create specific action plans, such as inventory management measures, new menu proposals, and sales promotion campaigns, based on the analysis results of the generative artificial intelligence model.

[1209] "Means for visually displaying proposals" refers to a dashboard or user interface that displays the generated proposals and analysis results in graphs or tables in a way that is easy for managers and staff to understand.

[1210] "Means for incorporating new data and retraining the generated AI model" refers to algorithms or software that use new data collected periodically to retrain the AI ​​model to improve its accuracy.

[1211] "Retail sales data" refers to data that includes transaction information when a product is sold at a retail store.

[1212] "Inventory data" refers to data that includes information indicating the inventory status of a retail store or restaurant at a specific point in time.

[1213] "Customer purchase history data" is data that includes records of products and services purchased by individual customers in the past.

[1214] "Customer order history data" refers to data in the food and beverage industry that includes records of food and drink orders that customers have made in the past.

[1215] "Trend data" is external data that shows consumption trends and fashions in a specific region or period, and is obtained from social media and local information services.

[1216] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries. The system utilizes generative artificial intelligence techniques to collect, preprocess, analyze, generate recommendations, and visually display data. The details of this system are described below.

[1217] The server uses APIs and communication modules to collect data from each retail and restaurant chain. For example, it obtains sales and inventory data from a supermarket's sales database, and customer order history from a restaurant's ordering system. It also collects social media data and local trend data in parallel, taking into account the influence of external factors.

[1218] The collected data is preprocessed by the server. This preprocessing includes filling in missing data, correcting inventory data, standardizing data formats, and removing duplicate data. Specifically, if there are missing values ​​in sales data, they are filled in based on other related data. Anomaly detection is performed using TensorFlow and PyTorch algorithms to maintain data quality.

[1219] The server then inputs the preprocessed data into generative artificial intelligence models to perform demand forecasting and customer preference analysis. This analysis uses natural language processing (NLP) models and machine learning algorithms. For example, a model is run to predict next week's demand based on past sales data and local weather data. Another model is run to extract customer purchasing patterns and predict the popularity of certain products.

[1220] Based on the analysis results, the server generates specific proposals, such as next week's order list and inventory management for supermarkets, and new menu development and promotion campaigns for restaurants, to ensure smooth supply chain management and maximize sales opportunities.

[1221] The terminals display visual suggestions to managers and staff, including order lists in graph and list format, sales forecasts, campaign suggestions, etc. For example, a supermarket manager can see next week's order list displayed on the terminal and make necessary adjustments accordingly.

[1222] The system improves its accuracy through continuous learning: as new data is collected, the server retrains the generative artificial intelligence model to provide more optimal recommendations, allowing the system to evolve over time and become even more effective.

[1223] Examples of specific prompts for generative AI models

[1224] The following sentence can be used as an example of a prompt sentence to be input to the generative artificial intelligence model:

[1225] Example prompt 1:

[1226] "Predict next week's demand for ice cream based on weekly sales data from the past year and local weather data."

[1227] Example prompt 2:

[1228] "Using customer purchase history and social media trend data from the past three months, please suggest new menu items that will be most popular next month."

[1229] This makes it possible to automatically generate specific and effective proposals for the retail and food service industries.

[1230] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1231] Step 1: Data collection

[1232] The server collects various data from each retail and restaurant chain. Specifically, the server accesses the supermarket's sales database via API to obtain sales and inventory data. It also obtains customer order history from the restaurant's ordering system, and uses social media APIs and local information service APIs to collect social media data and local trend data. The input is the various data received from the API, and the output is raw collected data that is stored in a central database within the server.

[1233] Step 2: Data Preprocessing

[1234] The server preprocesses the collected data. First, if there is missing data, it fills it in based on other relevant data. Next, it uses statistical methods and machine learning algorithms to detect and correct outliers. If the data format is different, it is converted into a consistent format, and if there is duplicate data, it is removed. The input is the raw data collected in step 1, and the output is the clean data after preprocessing.

[1235] Step 3: Data analysis

[1236] The server inputs the preprocessed data into a generative artificial intelligence model for further analysis. For example, it uses TensorFlow and PyTorch to run demand forecasting and customer preference analysis models. Specifically, it uses past sales data and weather data to predict future demand and analyze customer purchasing patterns. The input is the preprocessed data, and the output is the analysis results, which include predicted sales volume, a list of popular products, and trend analysis results.

[1237] Step 4: Proposal Generation

[1238] The server generates specific proposals based on the analysis results generated. For example, it provides a supermarket with an order list for the next week based on the results of demand forecasting, or creates a promotional campaign for a specific customer group based on the results of customer preference analysis. It also proposes new menu additions to restaurants based on regional trends. The input is the analysis results generated in step 3, and the output is specific proposals.

[1239] Step 5: View the results

[1240] The terminal visualizes and displays the proposals provided by the server. It displays sales forecasts in graph form, order lists in list form, and campaign proposals on a dashboard accessible to managers and staff. For example, a supermarket manager can check next week's order list on the terminal and make necessary adjustments. The input is the proposal generated in step 4, and the output is the visual representation of the proposal.

[1241] Step 6: Continuous learning

[1242] The server retrains the generative AI model using newly collected data, which improves the model's accuracy over time and enables it to provide more accurate and useful suggestions. The input is the latest newly collected data, and the output is a retrained, highly accurate generative AI model.

[1243] (Application example 1)

[1244] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1245] Traditional food supply chain management systems are limited to collecting and analyzing data within stores and warehouses, and do not optimize the delivery process. This makes it difficult to select efficient routes during delivery or monitor progress in real time, which can lead to placement losses and deterioration of storage quality. Furthermore, delays in demand forecasting and response, as well as difficulties in optimal ordering and inventory management, lead to lost sales opportunities and excess inventory.

[1246] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1247] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating proposals based on the analysis results, means for visually displaying the proposals, means for incorporating new data and retraining the generative AI model, and means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles. This enables real-time data collection and analysis during the delivery process, making it possible to propose optimal delivery routes and times, and to realize efficient ordering and inventory management based on demand forecasts.

[1248] "Data collection means" refers to means for collecting location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, and customer purchase history data from retail and food service businesses.

[1249] The "preprocessing means" refers to means for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[1250] A "generative AI model analysis means" is a means for analyzing pre-processed data using a generative AI model to forecast demand, customer preferences, optimize delivery routes, and predict delivery times.

[1251] The "proposal generation means" is a means for generating optimal delivery routes, delivery times, and order lists based on the analysis results of the generative AI model.

[1252] The "visual display means" is a means for displaying generated proposals and analysis results in a format that is easy for managers and staff to understand.

[1253] "Retraining methods" are methods for incorporating new collected data into a generative AI model and retraining it to improve the model's accuracy.

[1254] An "autonomous vehicle" is a vehicle that can drive autonomously and deliver goods to a destination.

[1255] This invention is a food supply chain management system that uses autonomous vehicles, and aims to achieve efficient and effective data collection, data preprocessing, analysis using generative AI models, proposal generation, and visual display of results.

[1256] First, the server collects data. Specifically, it obtains location information, delivery progress data, and warehouse temperature data from autonomous vehicles, as well as sales data, inventory data, customer purchase history data, and order history data from partner stores and restaurants. It also simultaneously obtains trend data from external sources such as social media.

[1257] The server then preprocesses the collected data. This includes filling in missing values, detecting and correcting outliers, standardizing data formats, and removing duplicate data. For example, if there are missing sales data, a method is used to fill in estimated values ​​using other related data.

[1258] The preprocessed data is passed to a generative AI model. The server uses the generative AI model to perform detailed analysis of the data. Specific analysis items include demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. This makes it possible to predict demand for the following week based on past data, or predict the popularity of specific products.

[1259] Based on the analysis results, the server generates specific proposals, such as routes and delivery schedules for autonomous vehicles, next week's order lists and inventory management strategies for supermarkets and restaurants, and even suggestions for developing new menus and promotional campaigns.

[1260] The generated proposals are visually displayed on the terminal. The proposals are displayed in graph and list format on the terminal, making them easy for managers and staff to understand and implement. For example, a supermarket manager can check the order list on the terminal and make any necessary adjustments.

[1261] The server retrains the generative AI model every time new data is collected, improving the accuracy of the system, which allows it to provide timely and more optimal recommendations.

[1262] Specific examples

[1263] For example, if the server predicts that demand for ice cream will increase in a particular region during the summer, it will combine past sales data with local weather data to predict ice cream demand for the following week. As a result, a suggestion to order twice the usual amount will be displayed on the terminal. It may also analyze customer purchase history data to determine that a particular ice cream (such as chocolate) is very popular, and suggest launching a promotional campaign using this as a main product.

[1264] Example prompt sentence:

[1265] "Design an algorithm to predict ice cream demand in a specific region during the summer and suggest the optimal order quantity."

[1266] "Consider sales data from the past three months to forecast next week's sales of essential items and generate an optimal order list."

[1267] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1268] Step 1:

[1269] The server collects data, specifically location information from autonomous vehicles, delivery progress data, and warehouse temperature data, as well as sales data, inventory data, customer purchase history data, order history data, and external social media trend data from partner stores and restaurants. This allows the current status of the entire supply chain to be collected as data.

[1270] Input: location information from self-driving vehicles, delivery progress data, warehouse temperature data, sales data from stores and restaurants, inventory data, purchase history data, order history data, SNS trend data

[1271] Output: Raw data collected

[1272] Step 2:

[1273] The server performs preprocessing on the collected data, including filling in missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are missing sales data, it fills in the estimated value based on other related data. This ensures data consistency and quality.

[1274] Input: Raw data collected

[1275] Output: Preprocessed data

[1276] Step 3:

[1277] The server inputs the preprocessed data into a generative AI model for detailed analysis. Specifically, it performs demand forecasting, customer preference analysis, delivery route optimization, and delivery time prediction. The generative AI model predicts future trends from past data and suggests optimal actions.

[1278] Input: Preprocessed data

[1279] Output: Analysis results

[1280] Step 4:

[1281] The server generates specific proposals based on the analysis results of the generative AI model, such as optimal delivery routes and schedules for autonomous vehicles, order lists and inventory management strategies for the following week, and proposals for the development of new menu items and sales promotion campaigns.

[1282] Input: Analysis results

[1283] Output: Specific proposals (optimal delivery routes and schedules, order lists, inventory management measures, proposals for developing new menu items, proposals for sales promotion campaigns)

[1284] Step 5:

[1285] The server visually displays the generated proposals on the terminal, where they are displayed in graph and list format, providing managers and staff with a format that is easy to understand. This allows managers and staff to intuitively understand the proposals and make any necessary adjustments or actions.

[1286] Input: Specific proposal

[1287] Output: Visual display (graphs, list of suggestions)

[1288] Step 6:

[1289] The server retrains the generative AI model every time new data is collected, improving the model's accuracy and its ability to provide better recommendations. The retrained model ensures that predictions and recommendations are always based on the most up-to-date data.

[1290] Input: New data

[1291] Output: Retrained generative AI model

[1292] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1293] This invention relates to a system that combines generative AI technology with an emotion engine that recognizes user emotions to efficiently manage food supply chains in the retail and restaurant industries. The system provides functions for data collection, preprocessing, analysis, recommendation generation, visual display, continuous learning, and emotion recognition.

[1294] Program processing flow

[1295] 1. Data Collection

[1296] The server collects data from each connected retail and food service establishment. For example, it obtains sales data, inventory data, and customer purchase and order histories from supermarkets and restaurants via API. It also simultaneously collects social media and local trend data and stores it as external environment data.

[1297] 2. Data Preprocessing

[1298] The server performs various preprocessing operations on the collected data, such as filling in missing data values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data. For example, if there are gaps in sales data, the server fills in estimated values ​​based on similar data from the past.

[1299] 3. Demand forecasting

[1300] The server then inputs the pre-processed data into a generative AI model to forecast demand for the following week, using historical sales data, inventory data, and regional trend data to generate detailed forecasts of demand for specific days and specific products.

[1301] 4. Customer preference analysis

[1302] The server analyzes customer purchase and order histories to identify which products and categories a particular customer prefers, allowing it to generate personalized marketing messages and recommend specific products.

[1303] 5. Trend Analysis

[1304] The server analyzes social media data and local trend data to identify products and categories that are likely to be popular now or in the near future. For example, if a particular dish is trending in a local area on social media, it determines whether that dish should be added to a restaurant's menu.

[1305] 6. Emotion recognition

[1306] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to determine whether they are expressing positive or negative emotions. For example, if a customer's posts are positive, it can suggest new product information or promotions to the customer.

[1307] 7. Emotion-based suggestion generation

[1308] The server generates customized marketing messages and new product suggestions based on the customer's emotional state analyzed by the emotion engine, and if the customer is dissatisfied, suggests special offers or services to resolve the problem.

[1309] 8. Results display

[1310] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can review the display and make adjustments as necessary.

[1311] 9. Continuous learning

[1312] The server retrains the generative AI model and emotion engine as new data is collected, allowing the system to become more accurate over time and capable of providing more optimal recommendations.

[1313] Specific examples

[1314] For example, a local supermarket predicts that demand for a particular product (e.g., ice cream) will increase during the summer. The server combines past sales data with information about local heat waves to predict that demand for ice cream next week will be twice as high as usual. As a result, the terminal displays a suggestion to order twice the usual amount. At the same time, an analysis of customer purchase history reveals that chocolate ice cream is particularly popular, so the server suggests that a promotional campaign be launched focusing on this product.

[1315] In addition, a restaurant may analyze social media data and discover that a particular dish (e.g., homemade pizza) is popular in the neighborhood. Based on this trend, the server suggests adding homemade pizza to the menu and running a social media campaign. The suggestion is displayed on the device, and the restaurant manager can introduce the new menu item and run the campaign accordingly.

[1316] Furthermore, the emotion engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the customer an invitation to a tasting event for the new menu. The suggestion is displayed on the terminal, and the restaurant manager sends the invitation message to the customer according to the suggestion.

[1317] In this way, by combining generative AI technology with emotion recognition, the system can help retail and food service businesses efficiently and effectively manage their food supply chains, increasing customer satisfaction and sales.

[1318] The processing flow will be explained below.

[1319] Program processing steps

[1320] Step 1: Data collection

[1321] The server collects various data from each retail and restaurant establishment. Specifically, it periodically obtains sales data, inventory data, customer purchase history data, and customer order history data via API. At the same time, it also collects social media and regional trend data and stores it as external environment data for analysis.

[1322] Step 2: Data Preprocessing

[1323] The server performs preprocessing on the collected data. During this process, it complements missing values ​​in the data, detects and corrects outliers, standardizes data formats, and removes duplicate data. For example, if there is a defect in sales data, it can infer the defect from other related data and complement it.

[1324] Step 3: Demand forecast

[1325] The server then inputs the pre-processed data into a generative AI model to forecast demand for the next week or month. Based on past sales data, inventory data, and regional trend data, the server makes detailed forecasts of demand for specific products and days.

[1326] Step 4: Customer preference analysis

[1327] The server analyzes customer purchase and order history data to identify which products and categories a particular customer prefers, and uses this information to generate personalized marketing messages promoting specific products and services.

[1328] Step 5: Trend analysis

[1329] The server analyzes social media data and regional trend data to identify products and categories that are likely to be popular now and in the near future. Based on this information, it makes suggestions for adding new products or running promotional campaigns, for example.

[1330] Step 6: Emotion Recognition

[1331] The server uses an emotion engine to analyze the customer's emotional state. It analyzes the customer's purchasing behavior and social media posts to identify positive and negative emotions. For example, if a customer has positive feelings about a new product, that emotion can be used to promote the new product.

[1332] Step 7: Sentiment-based suggestion generation

[1333] The server generates customized marketing messages and new product proposals based on the customer's emotional state analyzed by the emotion engine. If a customer is dissatisfied with a particular product or service, the server will suggest special offers or services to resolve the issue.

[1334] Step 8: View the results

[1335] The terminal visually displays the proposals received from the server to managers and staff. The displayed content includes order lists in graph and list format, sales forecasts, campaign proposals, and how to respond to specific customers based on the results of sentiment analysis. Managers can check the display and adjust and implement the proposals as necessary.

[1336] Step 9: Take Action

[1337] Users (store managers and staff) follow instructions on the terminal to actually place orders and carry out sales promotion activities, such as placing orders with suppliers based on the proposed order list and sending customized messages to specific customers based on the results of sentiment analysis.

[1338] Step 10: Continuous learning

[1339] The server retrains the generative AI model and sentiment engine based on new data collected, further improving accuracy over time, giving the system the ability to provide increasingly accurate suggestions.

[1340] These are the specific processing steps in the AI ​​Food Partner system that combines an emotion engine. This is expected to improve the efficiency of the food supply chain, improve customer satisfaction, and increase sales.

[1341] Example 2

[1342] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1343] Efficient management of food supply chains in the retail and restaurant industries requires the collection and analysis of a wide variety of data, and marketing that takes customer emotions into consideration is also required. Conventional systems have difficulty processing this data in a centralized manner, which can lead to insufficient demand forecasting and analysis of customer preferences. Furthermore, it is difficult to generate proposals that take customer emotions into consideration, leaving challenges in improving customer satisfaction and sales.

[1344] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating suggestions based on the analysis results, means for visually displaying the suggestions, means for incorporating new data and retraining the generative AI model, means for analyzing emotions using an emotion engine that recognizes the emotional state of customers, and means for generating customized suggestions based on the analyzed emotional state. This enables efficient management of the food supply chain and improves customer satisfaction and sales by providing suggestions based on customer preferences and emotions.

[1345] The "means of collecting data" refers to a mechanism for collecting sales data, inventory data, customer purchase history data, order history data, and external trend data and social media data from each retail and food service store.

[1346] The "preprocessing means" is a processing mechanism for completing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data from collected data.

[1347] A "means for analyzing using a generative AI model" is a mechanism that uses preprocessed data as input and analyzes and predicts the data using a specific algorithm.

[1348] The "means for generating proposals" is a mechanism for creating customized marketing messages and product proposals based on the analysis results and sentiment analysis results.

[1349] The "visual display means" refers to a display device and its control mechanism for visually presenting the generated proposals and analysis results to managers and staff.

[1350] A "means for retraining" is a mechanism for retraining a generative AI model to improve its accuracy each time new data is acquired.

[1351] An "emotion engine" is an analytical device that analyzes a customer's emotional state from text and other data and determines whether the emotion is positive, negative, or neutral.

[1352] The "means for generating customized offers" is a mechanism for generating offers and marketing messages that are suited to individual customers based on the analyzed emotional state.

[1353] MODE FOR CARRYING OUT THE INVENTION

[1354] This invention relates to a system for efficiently managing food supply chains in the retail and restaurant industries, improving customer satisfaction and sales. The system provides functions for data collection, data preprocessing, analysis using generative AI models, recommendation generation, visual display, continuous learning, emotion recognition, and customized recommendations.

[1355] Hardware and software used

[1356] server

[1357] The server is a computer that runs the following software:

[1358] Data collection software: Acquires data via API, collects data using web crawling technology, etc.

[1359] Data preprocessing software: Using data analysis tools such as Pandas and Scikit-learn, we perform tasks such as imputing missing values, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[1360] Generative AI models: Data analysis and demand forecasting, including time series forecasting models built using PyTorch and TensorFlow.

[1361] Sentiment recognition engine: Analyze customer sentiment using a sentiment analysis engine such as Google Cloud Natural Language API.

[1362] Retraining capabilities: Ability to retrain generative AI models and sentiment analysis engines with each information update.

[1363] Terminal

[1364] The terminal is a computer or tablet for running the following software:

[1365] Visual display software: Use libraries such as D3.js and Chart.js to visually display recommendations and analysis results.

[1366] Specific examples

[1367] For example, when a local supermarket needs to forecast demand for a particular product, the server collects past sales data, inventory data, and local trend data (such as past weather information and social media trend information). This data is then preprocessed and input into a generative AI model. If the result predicts that demand next week will be twice the usual amount, the device will visually display the detailed forecast along with a suggestion to double the order quantity.

[1368] Furthermore, when analyzing social media posts using emotion recognition, it can be determined that a particular dish (e.g., homemade pizza) is popular in a region. Based on this information, the server generates a proposal to add homemade pizza to the menu and run a campaign on social media. The proposal and specific campaign methods are visually displayed on the device.

[1369] Furthermore, the emotion recognition engine analyzes customer posts on social media to detect that a particular customer has very positive feelings about the new dessert menu. Based on this information, the server generates a suggestion to send the relevant customer an invitation message to a tasting event for the new menu. The suggestion is displayed on the device, and the user can send the message to the customer according to the suggestion.

[1370] Prompt Sentence Examples

[1371] Demand forecast prompt statement

[1372] python

[1373] past_sales_data = get_past_sales_data()

[1374] trend_data = get_trend_data()

[1375] input_data = combine_data(past_sales_data, trend_data)

[1376] predicted_demand = ai_model.predict(input_data)

[1377] print(predicted_demand)

[1378] Customer Preference Analysis Prompt

[1379] python

[1380] customer_purchase_history = get_customer_purchase_history(customer_id)

[1381] recommendations = recommend_products(customer_purchase_history)

[1382] print(recommendations)

[1383] As described above, by combining generative AI technology with emotion recognition, this invention can help retail and food service businesses efficiently and effectively manage their food supply chains, improving customer satisfaction and sales.

[1384] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1385] Program processing flow and specific explanation

[1386] Step 1: Data collection

[1387] The server collects data from each retail and restaurant establishment, including the following data:

[1388] Input: Sales data, inventory data, customer purchase history, order history, social media data, and regional trend data obtained using API and web crawling technology.

[1389] Data processing: Store data is obtained in real time through APIs, and social media and trend data is collected by web crawling.

[1390] Output: Various datasets stored in a database.

[1391] Specifically, the server retrieves supermarket sales data through an API call and executes a crawling script to collect social media trends.

[1392] Step 2: Data Preprocessing

[1393] The server performs pre-processing on the collected data.

[1394] Input: Raw data collected in step 1.

[1395] Data processing: Using Pandas and Scikit-learn, we perform missing value imputation, outlier detection and correction, data format standardization, and duplicate data removal.

[1396] Output: A preprocessed and clean dataset.

[1397] As a specific operation, for example, if there is a missing value in the sales data, the server will calculate the average value from similar past data and fill in the missing value.

[1398] Step 3: Demand forecast

[1399] The server uses a generative AI model to predict demand.

[1400] Input: The preprocessed dataset from step 2.

[1401] Data Computing: Use time series forecasting models to forecast demand for the next week. Use generative AI models in PyTorch and TensorFlow.

[1402] Output: Forecasted demand data.

[1403] Specifically, the server inputs past sales and inventory data into the generative AI model, obtaining a prediction result such as, "Demand for ice cream will be twice the usual amount next Tuesday."

[1404] Step 4: Customer preference analysis

[1405] The server analyzes the customer's purchasing history and other information to identify individual preferences.

[1406] Input: Customer purchase and order history data preprocessed in step 2.

[1407] Data computation: Using Python and Scikit-learn, data is clustered to identify customer preferences.

[1408] Output: A list of each customer's preferred products and categories.

[1409] Specifically, the server may identify, for example, that a customer has frequently purchased chocolate-related products over the past six months, and then use that information to suggest a special campaign.

[1410] Step 5: Trend analysis

[1411] The server analyzes social media data and regional trend data to identify trends.

[1412] Input: Social media data and local trend data collected in Step 1.

[1413] Data Computing: Use NLTK and Spacy to perform text analysis and identify trending products and categories.

[1414] Output: A list of trending items and categories.

[1415] Specifically, the server detects trends such as "homemade pizza is becoming a hot topic on local social media."

[1416] Step 6: Emotion Recognition

[1417] The server analyzes the customer's emotions using an emotion engine.

[1418] Input: Text data such as social media posts and reviews.

[1419] Data calculation: Uses Google Cloud Natural Language API to identify positive, negative, and neutral sentiment.

[1420] Output: Customer sentiment analysis data.

[1421] As a specific operation, the server obtains a result such as "There are many posts expressing very positive sentiment toward this new dessert menu."

[1422] Step 7: Sentiment-based suggestion generation

[1423] The server generates customized suggestions based on the emotional state analyzed by the emotion engine.

[1424] Input: Sentiment analysis result data from step 6.

[1425] Data Computing: Generate customized marketing messages and new product offers using Python and JavaScript.

[1426] Output: A customized suggestion message.

[1427] As a specific example of its operation, the server may make a suggestion such as "generating an invitation message to a tasting event of a new menu for customers who have expressed positive emotions."

[1428] Step 8: View the results

[1429] The terminal visually displays the suggestions received from the server to the manager or staff.

[1430] Input: Proposed data and analysis results received from the server.

[1431] Display: Display in graph or list format using D3.js or Chart.js.

[1432] Output: Visually displayed recommendations and analysis results.

[1433] Specifically, the device will display a message with a graph, such as, "Demand for ice cream will be twice the usual amount next Tuesday, so please double your order quantity."

[1434] Step 9: Continuous learning

[1435] The server takes in new data and retrains the generative AI model and emotion engine.

[1436] Input: The newly collected dataset.

[1437] Data computation: Update model parameters and retrain using TensorFlow or PyTorch.

[1438] Output: An improved generative AI model and emotion engine.

[1439] Specifically, as data is updated weekly, sales trends and changes in customer sentiment are reflected in the model, improving the accuracy of the next forecast.

[1440] (Application example 2)

[1441] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1442] The retail and restaurant industries require systems that efficiently manage food supply chains while taking into account customer emotions. However, current systems are unable to generate proposals that fully reflect customer emotions and preferences, posing challenges in improving customer satisfaction and making appropriate demand forecasts. Therefore, a method is needed that analyzes customers' emotional states using a variety of data and makes personalized proposals.

[1443] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1444] In this invention, the server includes means for collecting data, means for preprocessing the collected data, means for analyzing the preprocessed data with a generative AI model, means for generating recommendations based on the analysis results, means for visually displaying the recommendations, means for incorporating new data and retraining the generative AI model, means for analyzing user sentiment, and means for generating recommendations based on the sentiment analysis results, thereby enabling personalized food supply chain management that takes into account the emotional state of customers.

[1445] "Means for collecting data" refers to a mechanism for obtaining sales data, inventory data, customer purchasing history data, social media data, and regional trend data from each retail and food service store.

[1446] The "preprocessing means" is a mechanism for complementing missing values ​​in collected data, detecting and correcting outliers, standardizing data formats, and deleting duplicate data.

[1447] "Means for analysis using a generative AI model" refers to a mechanism for performing demand forecasting and customer preference analysis using a generative AI model based on preprocessed data.

[1448] The "means for generating proposals" is a mechanism for creating personalized marketing messages and new product proposals based on the analysis results and sentiment analysis results of the generative AI model.

[1449] The "visual display means" is a mechanism for visually displaying the proposals received from the server to the administrator or staff in the form of graphs, lists, etc.

[1450] A "means for retraining the generative AI model" is a mechanism for retraining the generative AI model each time new data is collected, improving its accuracy over time.

[1451] "Means for analyzing user emotions" refers to a mechanism that analyzes customer purchasing behavior and social media posting data to determine whether emotions are positive or negative.

[1452] The "means for generating proposals based on the results of sentiment analysis" is a mechanism for generating customized marketing messages and new product proposals for customers based on the results of sentiment analysis.

[1453] "Means for collecting retail sales data, inventory data, and customer purchase history data" refers to mechanisms such as APIs and sensors for obtaining sales, inventory, and customer purchase history from retail stores.

[1454] "Means for collecting food service sales data and customer order history data" refers to mechanisms such as APIs and sensors for obtaining sales and specific customer order histories from food service establishments.

[1455] "Means for collecting trend data from outside" refers to a mechanism for obtaining regional trend information from social media and the Internet.

[1456] The "means for imputing missing values ​​and detecting and correcting outliers" is a mechanism for estimating and imputing missing values, detecting outliers, and appropriately correcting them for collected data.

[1457] The "means for unifying data formats and deleting duplicate data" is a mechanism for unifying data of different formats and deleting duplicate data.

[1458] The "means for generating personalized proposals based on the results of sentiment analysis" is a mechanism for using the results of sentiment analysis to make personalized product and service proposals to individual customers.

[1459] MODE FOR CARRYING OUT THE INVENTION

[1460] This invention is a system for efficiently and effectively managing food supply chains in the retail and restaurant industries. The system combines generative AI technology with an emotion engine that recognizes user emotions to provide data collection, preprocessing, analysis, recommendation generation, visual display, and continuous learning functions. The system can also be implemented as a smartphone application. Specific embodiments of the system are described below.

[1461] Hardware and Software

[1462] Hardware

[1463] Smartphone (iOS or Android)

[1464] Software Framework

[1465] requests: for API calls

[1466] json: For working with JSON data

[1467] textblob: for sentiment analysis

[1468] sklearn: For data preprocessing and demand forecasting

[1469] pandas: for data preprocessing

[1470] Data flow

[1471] The server processes the data through the following steps:

[1472] 1. Data Collection

[1473] The server collects sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant establishments via API.

[1474] 2. Data Preprocessing

[1475] The server complements missing values ​​in the collected data, detects and corrects outliers, standardizes data formats, and deletes duplicate data.

[1476] 3. Demand forecasting

[1477] The server inputs the pre-processed data into a generative AI model to predict future demand.

[1478] 4. Emotion recognition

[1479] The server uses an emotion engine to analyze the user's emotional state and determine whether the emotion is positive or negative based on social media posts and purchase history data.

[1480] 5. Proposal generation

[1481] The server generates personalized marketing messages and new product suggestions based on the analysis results of the generative AI model and the results of sentiment analysis.

[1482] 6. Results display

[1483] The terminal visually displays the suggestions received from the server to managers and staff in the form of graphs and lists.

[1484] 7. Continuous learning

[1485] The server retrains the generative AI model and emotion engine every time new data is collected.

[1486] Specific examples

[1487] For example, suppose a user enters the text "I'm feeling really down today..." To parse this text, the server does the following:

[1488] 1. Emotion recognition

[1489] The server performs sentiment analysis using the TextBlob library and returns a negative sentiment score.

[1490] 2. Proposal generation

[1491] The server generates a suggestion, "Comfort food specially for you!" based on the negative sentiment score, past purchase history, and trend data.

[1492] The suggestions are visually displayed to the user in an app on their device, providing them with food delivery suggestions tailored to their emotional state that day.

[1493] Prompt Sentence Examples

[1494] An example of a prompt sentence to input to the generative AI model is:

[1495] "Create recommended food delivery suggestions based on the user's purchasing history and sentiment score. For example, if a user types, 'I am feeling so happy today!', provide suggestions based on positive sentiment."

[1496] Examples include:

[1497] In this way, the system of the present invention improves customer satisfaction and realizes efficient management of the food supply chain by providing personalized suggestions based on the user's emotional state.

[1498] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1499] Step 1:

[1500] Data collection

[1501] The server uses APIs to obtain sales data, inventory data, customer purchase history, order history, social media data, and regional trend data from retail and restaurant stores. This data is collected in JSON format. The input is the store's API key, and the output is a JSON-formatted dataset.

[1502] Step 2:

[1503] Data Preprocessing

[1504] The server performs missing value completion, outlier detection and correction, data format standardization, and duplicate data removal for the collected data. The input is the JSON dataset obtained in step 1. For each data entry, if a missing value exists, it is completed based on similar data, and outliers are corrected according to certain criteria. The output is a clean dataset.

[1505] Step 3:

[1506] Demand forecasting

[1507] The server inputs the preprocessed data into a generative AI model to predict future demand. The input is the preprocessed dataset, and the output is the demand forecast data for the next week. Specifically, it uses past sales data, inventory data, and regional trend data to predict demand for a specific day or specific product.

[1508] Step 4:

[1509] emotion recognition

[1510] The server uses an emotion engine to analyze the user's emotional state. It determines whether the emotion is positive or negative by analyzing social media posts and purchase history data. The input is social media posts and purchase history data, and the output is an emotion score.

[1511] Step 5:

[1512] Proposal generation

[1513] The server generates personalized marketing messages and new product proposals based on the analysis results of the generative AI model and the sentiment analysis results. The input is demand forecast data and sentiment scores, and the output is personalized proposals. Specifically, it prepares proposals with rewards for positive sentiment scores and includes consolation messages for negative sentiment scores.

[1514] Step 6:

[1515] Results display

[1516] The terminal visually displays the proposals received from the server in graph or list format to managers and staff. The input is the proposal data from the server, and the output is the visually displayed information. Specifically, it displays campaign proposals and customer support methods on a dashboard for managers.

[1517] Step 7:

[1518] Continuous Learning

[1519] The server retrains the generative AI model and emotion engine every time new data is collected. The input is a new dataset, and the output is a retrained model. Specifically, the model is periodically updated to improve its accuracy.

[1520] 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 control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1521] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1522] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1523] The emotion identification model 59 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 an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1524] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1525] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1526] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1527] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1528] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs 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 a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1529] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1530] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1531] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1532] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1533] 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.

[1534] It is not necessary to store all 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 all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1535] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1536] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with 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). Also, the hardware resource that executes the specific processing may be a single processor.

[1537] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1538] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1539] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1540] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1541] The following is further disclosed regarding the above embodiment.

[1542] (Claim 1)

[1543] a means of collecting data;

[1544] means for pre-processing the collected data;

[1545] A means for analyzing the preprocessed data using a generative AI model;

[1546] means for generating recommendations based on the analysis results;

[1547] means for visually displaying said suggestions;

[1548] means for incorporating new data to retrain the generative AI model;

[1549] A system including:

[1550] (Claim 2)

[1551] A means for collecting retail sales data, inventory data, and customer purchase history data;

[1552] A means for collecting sales data and customer order history data for a food and beverage business;

[1553] A means for collecting trend data from an external source;

[1554] The system of claim 1 further comprising:

[1555] (Claim 3)

[1556] a means of imputing missing values ​​and detecting and correcting outliers;

[1557] A means of standardizing data formats and removing duplicate data;

[1558] The system of claim 1 further comprising:

[1559] "Example 1"

[1560] (Claim 1)

[1561] a means of collecting data;

[1562] means for pre-processing the collected data;

[1563] means for analyzing the preprocessed data using a generative artificial intelligence model;

[1564] means for generating recommendations based on the analysis results;

[1565] means for visually displaying said suggestions;

[1566] means for incorporating new data to retrain said generative artificial intelligence model;

[1567] A system including:

[1568] (Claim 2)

[1569] A means for collecting retail sales data, inventory data, and customer purchase history data;

[1570] A means for collecting food and beverage sales data and customer order history data;

[1571] A means for collecting trend data from an external source;

[1572] The system of claim 1 further comprising:

[1573] (Claim 3)

[1574] a means of imputing missing values ​​and detecting and correcting outliers;

[1575] A means of standardizing data formats and removing duplicate data;

[1576] The system of claim 1 further comprising:

[1577] "Application Example 1"

[1578] (Claim 1)

[1579] a means of collecting data;

[1580] means for pre-processing the collected data;

[1581] A means for analyzing the preprocessed data using a generative AI model;

[1582] means for generating recommendations based on the analysis results;

[1583] means for visually displaying said suggestions;

[1584] means for incorporating new data to retrain the generative AI model;

[1585] The system further includes means for collecting location information, delivery progress data, and warehouse temperature data from the autonomous vehicle.

[1586] (Claim 2)

[1587] A means for collecting retail sales data, inventory data, and customer purchase history data;

[1588] A means for collecting sales data and customer order history data for a food and beverage business;

[1589] A means for collecting trend data from an external source;

[1590] 10. The system of claim 1, further comprising means for optimizing delivery routes and predicting delivery times for autonomous vehicles.

[1591] (Claim 3)

[1592] a means of imputing missing values ​​and detecting and correcting outliers;

[1593] A means of standardizing data formats and removing duplicate data;

[1594] The system of claim 1, further comprising means for performing demand forecasting, delivery route optimization, and delivery time prediction using a generative AI model.

[1595] "Example 2: Combining Emotion Engines"

[1596] (Claim 1)

[1597] a means of collecting data;

[1598] means for pre-processing the collected data;

[1599] A means for analyzing the preprocessed data using a generative AI model;

[1600] means for generating recommendations based on the analysis results;

[1601] means for visually displaying said suggestions;

[1602] means for incorporating new data to retrain the generative AI model;

[1603] means for analyzing emotions using an emotion engine that recognizes the emotional state of the customer;

[1604] means for generating customized suggestions based on the analyzed emotional state;

[1605] A system including:

[1606] (Claim 2)

[1607] A means for collecting retail sales data, inventory data, and customer purchase history data;

[1608] A means for collecting sales data and customer order history data for a food and beverage business;

[1609] A means for collecting trend data from an external source;

[1610] A means of collecting customer social media data,

[1611] The system of claim 1 further comprising:

[1612] (Claim 3)

[1613] a means of imputing missing values ​​and detecting and correcting outliers;

[1614] A means of standardizing data formats and removing duplicate data;

[1615] A means of analyzing customer purchase and order histories to identify preferred products;

[1616] A means of analyzing SNS data and identifying trending products;

[1617] The system of claim 1 further comprising:

[1618] "Application example 2 when combining emotion engines"

[1619] (Claim 1)

[1620] a means of collecting data;

[1621] means for pre-processing the collected data;

[1622] A means for analyzing the preprocessed data using a generative AI model;

[1623] means for generating recommendations based on the analysis results;

[1624] means for visually displaying said suggestions;

[1625] means for incorporating new data to retrain the generative AI model;

[1626] A means of analyzing user emotions,

[1627] means for generating suggestions based on the sentiment analysis results;

[1628] A system including:

[1629] (Claim 2)

[1630] A means for collecting retail sales data, inventory data, and customer purchase history data;

[1631] A means for collecting sales data and customer order history data for a food and beverage business;

[1632] A means for collecting trend data from an external source;

[1633] A means of collecting social media data to analyze customer sentiment,

[1634] The system of claim 1 further comprising:

[1635] (Claim 3)

[1636] a means of imputing missing values ​​and detecting and correcting outliers;

[1637] A means of standardizing data formats and removing duplicate data;

[1638] a means for generating personalized suggestions based on the sentiment analysis results;

[1639] The system of claim 1 further comprising: [Explanation of symbols]

[1640] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means of collecting data; means for pre-processing the collected data; A means for analyzing the preprocessed data using a generative AI model; means for generating recommendations based on the analysis results; means for visually displaying said suggestions; means for incorporating new data to retrain the generative AI model; A system including:

2. A means for collecting retail sales data, inventory data, and customer purchase history data; A means for collecting sales data and customer order history data for a food and beverage business; A means for collecting trend data from an external source; The system of claim 1 further comprising:

3. a means of imputing missing values ​​and detecting and correcting outliers; A means of standardizing data formats and removing duplicate data; The system of claim 1 further comprising:

Images