System

Abstract

A system is provided.SOLUTION: A system including means for collecting knowledge and experience of maintenance support collected from a veteran engineer as digital data, means for normalizing and tagging the digital data, means for training a generated AI model using the normalized and tagged data, means for integrating the trained generated AI model into a real-time support system, means for receiving a query from a user, means for analyzing the query from the user and generating an optimal support procedure, means for providing the generated support procedure to the user, and means for collecting feedback after maintenance support and retraining the generated 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] Currently, when it comes to maintaining equipment in stores, the knowledge and experience gained is dependent on each technician, and there is insufficient collection and sharing of information. This results in the problem of work being concentrated on specific experienced technicians, and it takes time to train new technicians. By solving this problem, there is a need to provide an environment where all technicians can perform maintenance work quickly and efficiently. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for collecting maintenance response knowledge and experience from veteran engineers as digital data and normalizing and tagging it, a means for training a generative AI model using the digital data, a means for integrating the trained generative AI model into a real-time support system, a means for accepting and analyzing inquiries from users and generating optimal response procedures, a means for providing the generated response procedures to users, and a means for collecting feedback after maintenance responses and retraining the generative AI model. This promotes the sharing of knowledge and experience among engineers, enabling anyone to perform maintenance responses quickly and efficiently.

[0006] "Digital data" means a data format that allows information to be stored, processed, and transmitted electronically.

[0007] "Normalization" is the process of converting data into a standard format for consistency and accuracy.

[0008] "Tagging" is the process of attaching relevant keywords or labels to data to make it easier to search and categorize.

[0009] A "generative AI model" is a collection of artificial intelligence algorithms and structures that learn from collected data and perform specified tasks.

[0010] A "real-time support system" is a system that provides users with the information and support they need in real time.

[0011] An "inquiry" is a form of communication in which a user requests information or assistance from a system.

[0012] "Analysis" is the process of examining data or queries in detail and extracting meaning.

[0013] A "procedure" is a specific step or method to follow to resolve a particular problem.

[0014] "Feedback" refers to the evaluations and opinions that users provide about the system and its response methods.

[0015] "Retraining" is the process of retraining an existing generative AI model using new data and feedback. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] This invention is a system for speeding up and streamlining maintenance of storefront equipment, utilizing a generative AI model to aggregate the knowledge and experience of veterans, enabling all staff to work as experts. The system includes the following main components:

[0038] 1. Digitization of knowledge and experience

[0039] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[0040] 2. Data preprocessing and formatting

[0041] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0042] 3. Training the generative AI model

[0043] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[0044] 4. Providing real-time support

[0045] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0046] 5. Present and implement response procedures

[0047] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0048] 6. Gather feedback and retrain

[0049] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model and retrains it, continuously improving the system's accuracy and efficiency.

[0050] Examples:

[0051] Example 1: Screen not displaying issue

[0052] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The user then takes action according to the instructions, and the terminal receives feedback that the problem has been resolved.

[0053] Example 2: Communication error problem

[0054] When a user encounters a communication error, they use their device to send the error details to the server. The server uses a generative AI model to generate steps such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The user then follows the steps and responds, receiving feedback on the final result.

[0055] As described above, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[0059] Step 2:

[0060] The server normalizes and tags the collected data, standardizing its format and adding relevant keywords and labels, a process that makes the data more searchable and useful.

[0061] Step 3:

[0062] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0063] Step 4:

[0064] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[0065] Step 5:

[0066] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[0067] Step 6:

[0068] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[0069] Step 7:

[0070] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[0071] Step 8:

[0072] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[0073] Step 9:

[0074] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[0075] Step 10:

[0076] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[0077] Step 11:

[0078] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and improve the accuracy and efficiency of the system.

[0079] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[0080] Example 1

[0081] 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."

[0082] In conventional maintenance response systems, the experience and knowledge of technicians was highly dependent on the individual, making it difficult for new technicians to respond quickly and accurately. Furthermore, there was a lack of a way to efficiently share the know-how of veteran technicians, making it an issue to streamline troubleshooting.

[0083] 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.

[0084] In this invention, the server includes a means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data, a means for normalizing and tagging the digital data, and a means for training a generative AI model using the normalized and tagged data. This allows maintenance response knowledge to be digitized and standardized, and efficiently shared with new technicians through the AI ​​model. It also supports real-time troubleshooting, improving the quality and efficiency of maintenance response.

[0085] A "veteran technician" is a maintenance specialist with many years of experience and advanced technical knowledge.

[0086] "Knowledge and experience in maintenance" refers to troubleshooting and problem-solving procedures and related information obtained during maintenance work.

[0087] "Digital data" refers to information such as text, images, and video that has been converted into a form that a computer can process.

[0088] "Normalization" is the process of putting data into a consistent format, which is done to ensure data integrity and consistency.

[0089] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[0090] A "generative AI model" is an artificial intelligence model that learns problem-solving procedures and efficient response methods based on collected data and automatically generates them.

[0091] A "real-time support system" is a system that allows new engineers and others to receive immediate support while performing maintenance work.

[0092] "Means for accepting inquiries" refers to an interface or device for receiving questions or problems from users.

[0093] "Means for generating response procedures" refers to methods and techniques for creating optimal solution procedures based on the content of inquiries received from users.

[0094] "Means for collecting feedback" refers to a method or system for collecting evaluations and result information from users after maintenance has been performed.

[0095] "Retraining" is the process of re-learning an AI model using new data collected to improve its accuracy and efficiency.

[0096] "Data preprocessing and formatting" refers to a series of procedures for converting collected data into a format suitable for analysis or training.

[0097] This invention is a system for speeding up and streamlining maintenance responses for storefront equipment. It utilizes generative AI models to aggregate the knowledge and experience of veterans, providing a mechanism for all staff to respond as experts. The system includes the following main components:

[0098] 1. Digitization of knowledge and experience

[0099] The server collects maintenance knowledge and experience from veteran technicians and stores it as digital data. Specifically, it includes information such as past trouble cases recorded by technicians, problem-solving procedures, and methods for identifying troubles. This digital data is stored in a database and used for later processing. The collected data is stored in text, image, and video formats.

[0100] 2. Data preprocessing and formatting

[0101] The server normalizes the collected digital data and compiles it into a unified format. For example, it converts trouble cases recorded in different formats into a unified text format. It also tags the data with relevant keywords and classification labels. This improves data searchability and enables more efficient use.

[0102] 3. Training the generative AI model

[0103] The server uses preprocessed digital data to train a generative AI model. Large amounts of historical data are used, particularly to learn troubleshooting procedures and efficient response methods. This AI model is then trained using machine learning algorithms, giving it the ability to flexibly respond to new problems.

[0104] 4. Providing real-time support

[0105] Users (engineers and new employees) receive support for maintenance responses through a dedicated application. When the user inputs the problem into the device, the device sends the input information to the server. The server analyzes the received information and generates the optimal response procedure using a generative AI model. The generated procedure is then sent to the device, which displays it to the user.

[0106] 5. Present and implement response procedures

[0107] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0108] 6. Gather feedback and retrain

[0109] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, and response efficiency. The device then sends this feedback to the server, which processes it and reflects it in the generated AI model. The AI ​​model is retrained based on the feedback data, thereby continuously improving the accuracy and efficiency of the entire system.

[0110] Examples:

[0111] Example 1: Screen not displaying issue

[0112] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input "screen not displaying" into the terminal. The server analyzes this information and uses a generative AI model to suggest next steps:

[0113] 1. Check the power cable connection.

[0114] 2. Check that the voltage is correct.

[0115] 3. Reboot your system.

[0116] The user responds accordingly and provides feedback to the terminal that the problem has been solved.

[0117] Example 2: Communication error problem

[0118] When a user encounters a communication error, they input "communication error" into their device. The server uses a generative AI model to suggest the following steps:

[0119] 1. Check the cable connections.

[0120] 2. Restart your router.

[0121] 3. Check the server settings.

[0122] The user follows the procedure and responds by providing feedback on the final result.

[0123] As described above, this invention improves the quality and efficiency of maintenance response by consolidating the knowledge and experience of veterans and providing support in real time.

[0124] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0125] Step 1:

[0126] The server collects maintenance knowledge and experience from veteran engineers. Specifically, it collects data such as trouble cases recorded by engineers in the past, problem-solving procedures, and identification methods, and stores it in a database. The input is notes and reports on the engineers' experience and knowledge, and the output is digital data in a unified format.

[0127] Step 2:

[0128] The server normalizes and tags the collected digital data. The input is the digital data collected in step 1, and normalization and tagging improve the consistency and searchability of the data. The output is normalized and tagged data. Specifically, data in different formats is converted into a unified format and keywords and classification labels are assigned.

[0129] Step 3:

[0130] The server trains the generative AI model using the preprocessed digital data. The input is the data processed in step 2, and the machine learning algorithm is used to train the AI ​​model. The output is the trained generative AI model, which can generate troubleshooting procedures.

[0131] Step 4:

[0132] The user inputs the problem into the terminal. The specific input is a symptom of the problem, such as "the screen is not displayed" or "communication error." The terminal receives this input and sends it to the server. The input is the user's problem, and the output is the data to be sent.

[0133] Step 5:

[0134] The server analyzes the problem sent by the user and generates the optimal response procedure. The input is the problem data sent in step 4, and the response procedure is automatically generated using a generative AI model. The output is the generated response procedure. Specifically, the problem data is analyzed, and procedures such as "checking the power cable connection" and "measuring voltage" are created using the AI ​​model.

[0135] Step 6:

[0136] The server sends the generated response procedure to the terminal. The input is the response procedure generated in step 5, and the output is the data to be sent to the terminal. The specific operation is to transfer the generated procedure to the terminal.

[0137] Step 7:

[0138] The terminal displays the response procedures it has received to the user. The input is the response procedure data received in step 6, and the output is the display to the user. Specifically, the response procedures are displayed in list format on the terminal screen.

[0139] Step 8:

[0140] The user performs maintenance actions by following the procedures displayed on the terminal. The input is the procedures displayed in step 7, and the output is the results of performing the maintenance actions. The specific actions are the user performing physical operations based on the procedures.

[0141] Step 9:

[0142] The user inputs the results of the maintenance work into the terminal. Specific inputs include "the problem has been resolved" or "an additional problem has occurred." The terminal sends this feedback to the server. The input is the user's feedback, and the output is the data to send it.

[0143] Step 10:

[0144] The server processes the feedback and retrains the generative AI model. The input is the feedback data sent in step 9, and the output is the retrained generative AI model. The specific operation is to analyze the feedback data and update the AI ​​model taking into account the new data.

[0145] (Application example 1)

[0146] 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."

[0147] On-site maintenance is particularly difficult for new employees and inexperienced technicians, and requires a fast and accurate response. In addition, the robots used in factories have complex structures, so a high level of specialized knowledge is required to respond when a problem occurs. For this reason, a system is needed that allows new technicians to respond quickly and at the same level as experienced technicians.

[0148] 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.

[0149] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran engineers as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance and retraining the generative AI model; means for providing information to a dedicated device for performing maintenance on robots operating in factories; and means for displaying the maintenance response procedures to the user using the dedicated device. This enables even novice engineers to perform robot maintenance quickly and accurately.

[0150] A "veteran engineer" is an engineer who has advanced specialized knowledge and practical skills based on many years of experience.

[0151] "Maintenance response" refers to work such as inspection, repair, and adjustment carried out to maintain the normal operation of equipment and systems.

[0152] "Digital data" refers to data handled by digital devices such as computers, and includes text, images, videos, etc.

[0153] "Normalization" is the process of converting data into a unified format.

[0154] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[0155] A "generative AI model" is an artificial intelligence model that is trained using machine learning algorithms based on collected data.

[0156] A "real-time support system" is a system designed to allow users to receive support in real time.

[0157] "User inquiries" refer to the act of a user reporting a question or problem regarding maintenance support to the system.

[0158] "Specialized equipment" refers to devices or equipment specifically used for this invention.

[0159] "Means provided to the user" refers to the method by which the system presents the user with the most appropriate response procedures and information.

[0160] "Feedback after maintenance" refers to evaluations and opinions provided by users after maintenance work has been completed.

[0161] "Retraining" is the process of retraining a previously trained AI model with new data to improve the model's accuracy or performance.

[0162] "Robots operating in factories" refers to automated robots used on factory production lines and in workshops.

[0163] "Maintenance response procedures" refer to the specific procedures and steps for performing maintenance work on equipment or systems.

[0164] This invention is a system that efficiently performs maintenance on robots operating in factories by aggregating the maintenance knowledge and experience collected from skilled technicians as digital data and using that data to train a generative AI model. This system includes the following main components:

[0165] 1. Digitization of knowledge and experience

[0166] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[0167] 2. Data preprocessing and formatting

[0168] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0169] 3. Training the generative AI model

[0170] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[0171] 4. Providing real-time support

[0172] Users (engineers and new employees) receive support during robot maintenance through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0173] 5. Present and implement response procedures

[0174] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0175] 6. Gather feedback and retrain

[0176] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model for retraining, thereby continuously improving the system's accuracy and efficiency.

[0177] Examples:

[0178] Example 1: Robot arm not moving

[0179] When a user (a new engineer) encounters a problem where the robot arm won't move, they input the details of the problem using a dedicated terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "check the actuator operation," and "check the system log," and provides these to the user via the terminal. The user then takes action according to these steps, and receives feedback to the terminal that the problem has been resolved.

[0180] Example prompt sentence:

[0181] "What is the best procedure to solve the problem of a stuck robot arm?"

[0182] In this way, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[0183] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0184] Step 1:

[0185] The server collects maintenance knowledge and experience from veteran engineers as digital data. Specifically, information such as past trouble cases, problem-solving procedures, and methods for identifying troubles is digitized as text data, images, and videos. The input here is the experience and knowledge of the engineers, which is converted into digital format. The output is maintenance knowledge data stored as a digital archive.

[0186] Step 2:

[0187] The server normalizes and tags the collected digital data. Normalization is the process of standardizing the data format, and tagging is the process of assigning relevant keywords and classification labels. The input is the digital data, which is normalized and tagged. The output is the unified format and tagged data.

[0188] Step 3:

[0189] The server trains the generative AI model using the preprocessed digital data. After the data preprocessing is complete, the AI ​​model is trained using a machine learning algorithm. The input is the preprocessed digital data, and the output is the trained generative AI model.

[0190] Step 4:

[0191] The server integrates the trained generative AI model into the real-time support system. The AI ​​model is incorporated into the entire system so that it can generate maintenance response procedures in real time. The input is the trained generative AI model, and the output is the generative AI model that has become part of the real-time support system.

[0192] Step 5:

[0193] The terminal accepts inquiries from users. The user inputs questions or problems related to maintenance response through the terminal (e.g., smart glasses). The input is the content of the user's inquiry. The output is the content of the inquiry being sent to the server.

[0194] Step 6:

[0195] The server analyzes the inquiry received from the user and generates the optimal response procedure based on the generative AI model. The input is the inquiry from the user, and the output is the response procedure. Specifically, the generative AI model automatically generates the procedure that is most appropriate for the trouble content.

[0196] Step 7:

[0197] The terminal displays the response procedure sent from the server to the user. The input is the response procedure sent from the server, and the output is the specific maintenance response procedure provided to the user. Specifically, the procedure is displayed on the display of the smart glasses.

[0198] Step 8:

[0199] The user actually performs the maintenance work according to the procedure. During the process, the user carries out the necessary checks and repairs based on the procedure presented. The input is the procedure provided from the terminal, and the output is the result of the maintenance work.

[0200] Step 9:

[0201] After the maintenance response is completed, the user inputs feedback on the response results, required time, and response efficiency into the terminal. The input is feedback on the results and efficiency of the maintenance response, and the output is feedback data sent from the terminal to the server.

[0202] Step 10:

[0203] The server uses the collected feedback data to retrain the generative AI model: the input is the feedback data, and the output is a new generative AI model with improved accuracy and efficiency.

[0204] As a specific example, based on the prompt, "Please tell me the best procedure to solve the problem of the robot arm not moving," the user queries the server, and the server generates a procedure for the user to check using smart glasses while carrying out maintenance work.

[0205] 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.

[0206] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system for speeding up and streamlining maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[0207] 1. Digitization of knowledge and experience

[0208] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[0209] 2. Data Preprocessing and Normalization

[0210] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0211] 3. Training the generative AI model

[0212] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0213] 4. Providing real-time support

[0214] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0215] 5. Use of Emotion Engine

[0216] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0217] 6. Emotion-based adjustment of response procedures

[0218] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[0219] 7. Present and implement response procedures

[0220] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[0221] 8. Gather feedback and retrain

[0222] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[0223] Examples:

[0224] Example 1: Screen not displaying issue

[0225] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[0226] Example 2: Communication error problem

[0227] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[0228] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[0229] The processing flow will be explained below.

[0230] Step 1:

[0231] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[0232] Step 2:

[0233] The server normalizes and tags the collected data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0234] Step 3:

[0235] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0236] Step 4:

[0237] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[0238] Step 5:

[0239] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[0240] Step 6:

[0241] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[0242] Step 7:

[0243] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[0244] Step 8:

[0245] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[0246] Step 9:

[0247] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0248] Step 10:

[0249] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[0250] Step 11:

[0251] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[0252] Step 12:

[0253] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[0254] Step 13:

[0255] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[0256] Step 14:

[0257] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and emotion engine, improving the accuracy and efficiency of the system.

[0258] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[0259] Example 2

[0260] 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."

[0261] In on-site maintenance work, new and inexperienced engineers are required to solve problems quickly and efficiently on the spot. However, with conventional systems, it is difficult to share the knowledge and experience gained from veteran engineers in real time, and it is also difficult to respond flexibly while taking into account the user's emotional state. As a result, the quality and efficiency of maintenance work declines.

[0262] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0263] In this invention, the server includes: means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the user inquiries and generating optimal response procedures; means for providing the generated response procedures to the user; means for incorporating an emotion engine into the terminal and recognizing the emotional state of the user from facial expressions and voice; means for adjusting the response procedures based on data from the emotion engine; and means for collecting feedback after maintenance responses and retraining the generative AI model. This enables effective maintenance responses that take the user's emotional state into consideration, thereby improving the quality and efficiency of responses.

[0264] A "technician" is a person who has specialized knowledge and skills related to the maintenance, repair, or operation of a system or equipment.

[0265] "Digital data" refers to maintenance knowledge and experience expressed in an electronic format, and includes text data, images, audio data, and the like.

[0266] "Normalization" refers to the process of converting collected data into a unified format and making it consistent.

[0267] "Tagging" is the process of assigning relevant keywords and classification labels to digital data to make it easier to search and analyze later.

[0268] A "generative AI model" is an artificial intelligence model that can learn from collected and preprocessed data to perform specific tasks.

[0269] "Training" is the process of feeding data and adjusting the parameters of a generative AI model in order to learn useful knowledge and patterns.

[0270] A "real-time support system" is a system that provides immediate support for problems that arise during maintenance.

[0271] "User" refers to the engineers and technicians who use the system to perform maintenance.

[0272] The "emotion engine" is a system that analyzes the user's emotional state from their facial expressions and voice, and determines the appropriate response based on that.

[0273] "Feedback" refers to information about the results and emotional state of a user after a maintenance response.

[0274] "Retraining" is the process of retraining a generative AI model based on newly collected feedback data to improve the model's accuracy and performance.

[0275] The present invention relates to a system for quickly and effectively carrying out maintenance responses, and this system is mainly composed of three elements: a server, a terminal, and a user. How this system functions will be explained in detail below.

[0276] 1. Digitization of knowledge and experience

[0277] The server collects the maintenance knowledge and experience of veteran technicians as digital data. This includes text files, images, and audio data. For example, it collects examples of problems that technicians have solved in the past, along with detailed troubleshooting procedures. This data is stored in a database and can be used for later analysis and retrieval.

[0278] 2. Data Preprocessing and Normalization

[0279] The server converts the collected data into a unified format (normalizes it) and attaches relevant tags to it. Normalization and tagging improve the searchability and usability of the data. For example, adding tags such as "power supply trouble" or "network error" makes it easier to search later.

[0280] 3. Training the generative AI model

[0281] The server uses the preprocessed data to train the generative AI model. During this training process, a large amount of data is fed into the AI ​​model, and by adjusting the weights of the neural network, the AI ​​model learns how to resolve problems. For example, each maintenance procedure and solution method is memorized in the model.

[0282] 4. Providing real-time support

[0283] Users receive support during maintenance work using a dedicated application. The device accepts user input and sends it to a server. The server analyzes the received inquiry and generates optimal response procedures based on a generative AI model. The device then displays the generated response procedures to the user. For example, for a problem such as "the screen is not displaying," specific procedures such as "check the power cable connection" and "restart the system" are provided.

[0284] 5. Use of Emotion Engine

[0285] The device is equipped with an emotion engine that recognizes the user's emotional state from their facial expressions, voice, text, etc. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice and extract information such as whether the user is feeling stressed or relieved.

[0286] 6. Emotion-based adjustment of response procedures

[0287] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, it adds more detailed and careful explanations. In this way, the response procedures can be dynamically adjusted according to the user's emotional state.

[0288] 7. Present and implement response procedures

[0289] The terminal displays the adjusted response procedures to the user. For example, specific steps such as "check the power cable connection" and "measure the voltage" are displayed in a list, and the user follows these to perform the maintenance work.

[0290] 8. Gather feedback and retrain

[0291] After the maintenance response is completed, the user enters feedback into the device about the results, time required, response efficiency, and emotional state. The device then sends the feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine. This continuously improves the accuracy and efficiency of the system.

[0292] Examples:

[0293] Example 1: When a user encounters a problem with a device in a store, such as "the screen is not displaying," they input the details of the problem through the device. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection," "measure the voltage," and "restart the system." The device's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and responds, and the device receives feedback that the problem has been resolved.

[0294] Example 2: When a user encounters a "communication error," they use their device to send the error details to the server. The server uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to allow the user to complete the steps more quickly.

[0295] Example prompt sentence:

[0296] Sample prompt 1: "Please describe a problem a junior engineer encounters while working on a maintenance project in-store. Also, describe how the user's emotional state is reflected when the steps to resolve the problem are displayed. For example, when an error message appears."

[0297] Such a system not only improves the quality and efficiency of maintenance responses, but also enables flexible responses that take into account the user's emotional state.

[0298] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0299] System processing flow

[0300] Step 1: Gathering knowledge and experience

[0301] Input: Knowledge and experience from veteran engineers (text files, images, audio data, etc.)

[0302] The server collects maintenance knowledge and experience from veteran technicians, specifically documents, logs, voice memos, manuals, and other data used by technicians in maintenance operations, as digital data.

[0303] How it works: The server automatically retrieves files of various formats uploaded by technicians and stores them in a database, for example, by importing data from the technician's PC or cloud storage.

[0304] Output: Knowledge and experience in digital data format is stored in a database.

[0305] Step 2: Preprocessing and normalizing data

[0306] Input: Knowledge and experience of digital data formats

[0307] The server normalizes and tags the collected digital data, converting each piece of data into a unified format and assigning relevant keywords and classification labels.

[0308] How it works: The server converts, for example, audio data into text, parses the text data using natural language processing (NLP) to put it into a consistent format, and then assigns tags like "power outage" or "network error."

[0309] Output: A normalized and tagged dataset.

[0310] Step 3: Training the generative AI model

[0311] Input: Normalized and tagged dataset

[0312] The server uses the preprocessed data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods.

[0313] How it works: The server uses a neural network library (such as TensorFlow or PyTorch) to feed the tagged data to a generative AI model, adjusting the model's parameters (weights) to learn the optimal solution procedure.

[0314] Output: A trained generative AI model.

[0315] Step 4: Receiving inquiries from users

[0316] Input: User-entered problem details for maintenance

[0317] The user uses a dedicated application to input details of the problem that is currently being handled by the maintenance team. The terminal accepts this input and sends it to the server.

[0318] How it works: The user enters a specific problem into the application, such as "no screen display," and the device sends that information to the server.

[0319] Output: The query is sent to the server.

[0320] Step 5: Inquiry analysis and response procedure generation

[0321] Input: Inquiry sent by the user

[0322] The server analyzes the received inquiry and generates the optimal response procedure based on the generative AI model.

[0323] How it works: The server uses natural language processing to analyze the inquiry and generate specific response procedures such as "check the power cable connection" and "restart the system."

[0324] Output: The optimal response procedure is generated.

[0325] Step 6: View response procedures

[0326] Input: Generated response procedure

[0327] The terminal displays the generated response procedure to the user.

[0328] What it does: The device application will display instructions in text format, such as "Check the power cable. Then measure the voltage."

[0329] Output: Instructions for action are displayed to the user.

[0330] Step 7: Use the Emotion Engine

[0331] Input: User's facial expression and voice data

[0332] The terminal uses an emotion engine to recognize the user's emotional state from their facial expressions and voice.

[0333] How it works: The device analyzes images and audio captured by the camera and microphone to determine the user's emotions. For example, if the user is frowning, it will determine that the user is in a state of stress.

[0334] Output: User's emotional state data.

[0335] Step 8: Adjust response procedures

[0336] Input: User emotional state data

[0337] The server adjusts the response procedures generated by the generative AI model based on the emotional data from the emotion engine.

[0338] What it does: The server adds reassuring statements like "Don't worry" to the steps if the user is feeling anxious, or simplifies the steps if the user is feeling impatient.

[0339] Output: Coordinated response procedures.

[0340] Step 9: View the adjusted steps

[0341] Input: Coordinated response procedures

[0342] The terminal displays the adjusted procedure to the user.

[0343] What it does: The adjusted steps are displayed on the device, and the user can refer to them to perform the maintenance action. For example, the message "Don't worry, this step will be over soon."

[0344] Output: Adjusted response instructions displayed to the user.

[0345] Step 10: Gather feedback

[0346] Inputs: Maintenance response results, response time, response efficiency, emotional state

[0347] After the maintenance is completed, the user inputs feedback into the terminal.

[0348] Action: The user enters feedback such as "Problem solved", "Took 15 minutes", and "I was anxious during the process".

[0349] Output: Feedback data is sent from the device to the server.

[0350] Step 11: Retraining the generative AI model and emotion engine

[0351] Input: Collected feedback data

[0352] The server retrains the generative AI model and emotion engine based on newly collected feedback data.

[0353] How it works: The server uses the feedback data to readjust the parameters of the generative AI model to improve the accuracy and efficiency of the system, and also retrains the emotion engine to more accurately judge user emotions.

[0354] Output: Retrained generative AI model and emotion engine.

[0355] (Application example 2)

[0356] 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."

[0357] Existing maintenance support systems rely on the knowledge and experience of the technician, making it difficult for new or inexperienced technicians to provide effective maintenance support. Furthermore, support that does not take into account the user's emotional state can cause stress and anxiety. Another issue is the lack of real-time support, which means it takes a long time to resolve problems.

[0358] 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.

[0359] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance responses and retraining the generative AI model; means for adjusting the generated response procedures using an emotion engine that recognizes the user's emotions; and means for collecting and analyzing the user's emotions using a camera and microphone of a smart device, thereby enabling flexible and effective real-time support according to the user's emotional state.

[0360] "Collecting knowledge and experience" is the process of collecting knowledge and experience regarding maintenance responses from veteran technicians as digital data.

[0361] "Normalization and tagging" is the process of converting collected digital data into a unified format and assigning relevant keywords and classification labels.

[0362] "Training a generative AI model" is the process of using normalized and tagged data to teach a generative AI model how to respond efficiently.

[0363] "Real-time support system integration" is the process of building a system that uses trained generative AI models to provide support to users in real time.

[0364] "Accepting inquiries from users" is a means by which users can send questions about maintenance responses or trouble reports to the system.

[0365] "Analysis of inquiries and generation of response procedures" is a process of analyzing an inquiry from a user and generating an optimal response procedure based on that analysis.

[0366] An "emotion engine" is an engine that recognizes and analyzes a user's emotional state from facial expressions, voice, text, etc.

[0367] "Use of smart devices" refers to the use of devices such as smartphones and tablets to collect the user's emotional state through cameras and microphones.

[0368] "Feedback collection and retraining" refers to the process of collecting feedback from users after maintenance is performed and retraining the generative AI model based on that feedback.

[0369] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system that speeds up and streamlines maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[0370] 1. Digitization of knowledge and experience

[0371] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[0372] 2. Data Preprocessing and Normalization

[0373] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0374] 3. Training the generative AI model

[0375] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0376] 4. Providing real-time support

[0377] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0378] 5. Use of Emotion Engine

[0379] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0380] 6. Emotion-based adjustment of response procedures

[0381] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[0382] 7. Present and implement response procedures

[0383] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[0384] 8. Gather feedback and retrain

[0385] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[0386] Examples:

[0387] Example 1: Screen not displaying issue

[0388] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[0389] Example 2: Communication error problem

[0390] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[0391] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[0392] Example prompt sentence:

[0393] If the user types "screen not showing":

[0394] User is feeling anxious. Provide detailed and comforting support message. Base message: "Please check the connection of the power cable of the screen."

[0395] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0396] Step 1:

[0397] The terminal accepts the contents of a maintenance inquiry from the user as input. The user inputs the inquiry into the terminal, and the terminal transmits the inquiry contents to the server as text data.

[0398] Step 2:

[0399] The server analyzes the inquiry received from the device. The server uses a generative AI model to analyze the inquiry and generate appropriate response procedures. The server uses the text data of the inquiry as input and provides the generated text of the response procedures as output.

[0400] Step 3:

[0401] The terminal provides the user with the response procedure sent from the server, and receives the generated response procedure text from the server and displays it to the user in an appropriate format.

[0402] Step 4:

[0403] The device uses an emotion engine to collect the user's emotional state as input. It uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to obtain emotion data. It uses camera images and audio data as input and obtains emotional state data as output.

[0404] Step 5:

[0405] The server adjusts the response procedure created by the generative AI model based on the acquired emotional state data. It receives the emotional state data as input and inputs the prompt sentence into the generative AI model to create a newly adjusted response procedure. As an output, it obtains the text of the response procedure adjusted based on the emotion.

[0406] Step 6:

[0407] The terminal displays the adjusted response procedure to the user. The terminal receives the text of the adjusted response procedure from the server and displays it to the user.

[0408] Step 7:

[0409] The user performs maintenance and inputs the results as feedback to the terminal. After the maintenance is completed, the results, required time, emotional state, etc. are input to the terminal, and the terminal sends the contents to the server. The text data of the feedback content is used as input, and feedback data is provided to be sent to the server as output.

[0410] Step 8:

[0411] The server retrains the generative AI model using the received feedback data. It receives the feedback data as input, adjusts the weights of the generative AI model based on the feedback data, and retrains the model. It provides an updated generative AI model as output.

[0412] 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.

[0413] 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.

[0414] 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.

[0415] [Second embodiment]

[0416] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0417] 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.

[0418] 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).

[0419] 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.

[0420] 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.

[0421] 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).

[0422] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0423] 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.

[0424] 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.

[0425] 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.

[0426] 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.

[0427] 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."

[0428] This invention is a system for speeding up and streamlining maintenance of storefront equipment, utilizing a generative AI model to aggregate the knowledge and experience of veterans, enabling all staff to work as experts. The system includes the following main components:

[0429] 1. Digitization of knowledge and experience

[0430] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[0431] 2. Data preprocessing and formatting

[0432] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0433] 3. Training the generative AI model

[0434] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[0435] 4. Providing real-time support

[0436] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0437] 5. Present and implement response procedures

[0438] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0439] 6. Gather feedback and retrain

[0440] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model and retrains it, continuously improving the system's accuracy and efficiency.

[0441] Examples:

[0442] Example 1: Screen not displaying issue

[0443] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The user then takes action according to the instructions, and the terminal receives feedback that the problem has been resolved.

[0444] Example 2: Communication error problem

[0445] When a user encounters a communication error, they use their device to send the error details to the server. The server uses a generative AI model to generate steps such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The user then follows the steps and responds, receiving feedback on the final result.

[0446] As described above, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[0447] The processing flow will be explained below.

[0448] Step 1:

[0449] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[0450] Step 2:

[0451] The server normalizes and tags the collected data, standardizing its format and adding relevant keywords and labels, a process that makes the data more searchable and useful.

[0452] Step 3:

[0453] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0454] Step 4:

[0455] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[0456] Step 5:

[0457] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[0458] Step 6:

[0459] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[0460] Step 7:

[0461] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[0462] Step 8:

[0463] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[0464] Step 9:

[0465] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[0466] Step 10:

[0467] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[0468] Step 11:

[0469] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and improve the accuracy and efficiency of the system.

[0470] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[0471] Example 1

[0472] 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."

[0473] In conventional maintenance response systems, the experience and knowledge of technicians was highly dependent on the individual, making it difficult for new technicians to respond quickly and accurately. Furthermore, there was a lack of a way to efficiently share the know-how of veteran technicians, making it an issue to streamline troubleshooting.

[0474] 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.

[0475] In this invention, the server includes a means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data, a means for normalizing and tagging the digital data, and a means for training a generative AI model using the normalized and tagged data. This allows maintenance response knowledge to be digitized and standardized, and efficiently shared with new technicians through the AI ​​model. It also supports real-time troubleshooting, improving the quality and efficiency of maintenance response.

[0476] A "veteran technician" is a maintenance specialist with many years of experience and advanced technical knowledge.

[0477] "Knowledge and experience in maintenance" refers to troubleshooting and problem-solving procedures and related information obtained during maintenance work.

[0478] "Digital data" refers to information such as text, images, and video that has been converted into a form that a computer can process.

[0479] "Normalization" is the process of putting data into a consistent format, which is done to ensure data integrity and consistency.

[0480] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[0481] A "generative AI model" is an artificial intelligence model that learns problem-solving procedures and efficient response methods based on collected data and automatically generates them.

[0482] A "real-time support system" is a system that allows new engineers and others to receive immediate support while performing maintenance work.

[0483] "Means for accepting inquiries" refers to an interface or device for receiving questions or problems from users.

[0484] "Means for generating response procedures" refers to methods and techniques for creating optimal solution procedures based on the content of inquiries received from users.

[0485] "Means for collecting feedback" refers to a method or system for collecting evaluations and result information from users after maintenance has been performed.

[0486] "Retraining" is the process of re-learning an AI model using new data collected to improve its accuracy and efficiency.

[0487] "Data preprocessing and formatting" refers to a series of procedures for converting collected data into a format suitable for analysis or training.

[0488] This invention is a system for speeding up and streamlining maintenance responses for storefront equipment. It utilizes generative AI models to aggregate the knowledge and experience of veterans, providing a mechanism for all staff to respond as experts. The system includes the following main components:

[0489] 1. Digitization of knowledge and experience

[0490] The server collects maintenance knowledge and experience from veteran technicians and stores it as digital data. Specifically, it includes information such as past trouble cases recorded by technicians, problem-solving procedures, and methods for identifying troubles. This digital data is stored in a database and used for later processing. The collected data is stored in text, image, and video formats.

[0491] 2. Data preprocessing and formatting

[0492] The server normalizes the collected digital data and compiles it into a unified format. For example, it converts trouble cases recorded in different formats into a unified text format. It also tags the data with relevant keywords and classification labels. This improves data searchability and enables more efficient use.

[0493] 3. Training the generative AI model

[0494] The server uses preprocessed digital data to train a generative AI model. Large amounts of historical data are used, particularly to learn troubleshooting procedures and efficient response methods. This AI model is then trained using machine learning algorithms, giving it the ability to flexibly respond to new problems.

[0495] 4. Providing real-time support

[0496] Users (engineers and new employees) receive support for maintenance responses through a dedicated application. When the user inputs the problem into the device, the device sends the input information to the server. The server analyzes the received information and generates the optimal response procedure using a generative AI model. The generated procedure is then sent to the device, which displays it to the user.

[0497] 5. Present and implement response procedures

[0498] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0499] 6. Gather feedback and retrain

[0500] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, and response efficiency. The device then sends this feedback to the server, which processes it and reflects it in the generated AI model. The AI ​​model is retrained based on the feedback data, thereby continuously improving the accuracy and efficiency of the entire system.

[0501] Examples:

[0502] Example 1: Screen not displaying issue

[0503] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input "screen not displaying" into the terminal. The server analyzes this information and uses a generative AI model to suggest next steps:

[0504] 1. Check the power cable connection.

[0505] 2. Check that the voltage is correct.

[0506] 3. Reboot your system.

[0507] The user responds accordingly and provides feedback to the terminal that the problem has been solved.

[0508] Example 2: Communication error problem

[0509] When a user encounters a communication error, they input "communication error" into their device. The server uses a generative AI model to suggest the following steps:

[0510] 1. Check the cable connections.

[0511] 2. Restart your router.

[0512] 3. Check the server settings.

[0513] The user follows the procedure and responds by providing feedback on the final result.

[0514] As described above, this invention improves the quality and efficiency of maintenance response by consolidating the knowledge and experience of veterans and providing support in real time.

[0515] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0516] Step 1:

[0517] The server collects maintenance knowledge and experience from veteran engineers. Specifically, it collects data such as trouble cases recorded by engineers in the past, problem-solving procedures, and identification methods, and stores it in a database. The input is notes and reports on the engineers' experience and knowledge, and the output is digital data in a unified format.

[0518] Step 2:

[0519] The server normalizes and tags the collected digital data. The input is the digital data collected in step 1, and normalization and tagging improve the consistency and searchability of the data. The output is normalized and tagged data. Specifically, data in different formats is converted into a unified format and keywords and classification labels are assigned.

[0520] Step 3:

[0521] The server trains the generative AI model using the preprocessed digital data. The input is the data processed in step 2, and the machine learning algorithm is used to train the AI ​​model. The output is the trained generative AI model, which can generate troubleshooting procedures.

[0522] Step 4:

[0523] The user inputs the problem into the terminal. The specific input is a symptom of the problem, such as "the screen is not displayed" or "communication error." The terminal receives this input and sends it to the server. The input is the user's problem, and the output is the data to be sent.

[0524] Step 5:

[0525] The server analyzes the problem sent by the user and generates the optimal response procedure. The input is the problem data sent in step 4, and the response procedure is automatically generated using a generative AI model. The output is the generated response procedure. Specifically, the problem data is analyzed, and procedures such as "checking the power cable connection" and "measuring voltage" are created using the AI ​​model.

[0526] Step 6:

[0527] The server sends the generated response procedure to the terminal. The input is the response procedure generated in step 5, and the output is the data to be sent to the terminal. The specific operation is to transfer the generated procedure to the terminal.

[0528] Step 7:

[0529] The terminal displays the response procedures it has received to the user. The input is the response procedure data received in step 6, and the output is the display to the user. Specifically, the response procedures are displayed in list format on the terminal screen.

[0530] Step 8:

[0531] The user performs maintenance actions by following the procedures displayed on the terminal. The input is the procedures displayed in step 7, and the output is the results of performing the maintenance actions. The specific actions are the user performing physical operations based on the procedures.

[0532] Step 9:

[0533] The user inputs the results of the maintenance work into the terminal. Specific inputs include "the problem has been resolved" or "an additional problem has occurred." The terminal sends this feedback to the server. The input is the user's feedback, and the output is the data to send it.

[0534] Step 10:

[0535] The server processes the feedback and retrains the generative AI model. The input is the feedback data sent in step 9, and the output is the retrained generative AI model. The specific operation is to analyze the feedback data and update the AI ​​model taking into account the new data.

[0536] (Application example 1)

[0537] 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."

[0538] On-site maintenance is particularly difficult for new employees and inexperienced technicians, and requires a fast and accurate response. In addition, the robots used in factories have complex structures, so a high level of specialized knowledge is required to respond when a problem occurs. For this reason, a system is needed that allows new technicians to respond quickly and at the same level as experienced technicians.

[0539] 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.

[0540] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran engineers as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance and retraining the generative AI model; means for providing information to a dedicated device for performing maintenance on robots operating in factories; and means for displaying the maintenance response procedures to the user using the dedicated device. This enables even novice engineers to perform robot maintenance quickly and accurately.

[0541] A "veteran engineer" is an engineer who has advanced specialized knowledge and practical skills based on many years of experience.

[0542] "Maintenance response" refers to work such as inspection, repair, and adjustment carried out to maintain the normal operation of equipment and systems.

[0543] "Digital data" refers to data handled by digital devices such as computers, and includes text, images, videos, etc.

[0544] "Normalization" is the process of converting data into a unified format.

[0545] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[0546] A "generative AI model" is an artificial intelligence model that is trained using machine learning algorithms based on collected data.

[0547] A "real-time support system" is a system designed to allow users to receive support in real time.

[0548] "User inquiries" refer to the act of a user reporting a question or problem regarding maintenance support to the system.

[0549] "Specialized equipment" refers to devices or equipment specifically used for this invention.

[0550] "Means provided to the user" refers to the method by which the system presents the user with the most appropriate response procedures and information.

[0551] "Feedback after maintenance" refers to evaluations and opinions provided by users after maintenance work has been completed.

[0552] "Retraining" is the process of retraining a previously trained AI model with new data to improve the model's accuracy or performance.

[0553] "Robots operating in factories" refers to automated robots used on factory production lines and in workshops.

[0554] "Maintenance response procedures" refer to the specific procedures and steps for performing maintenance work on equipment or systems.

[0555] This invention is a system that efficiently performs maintenance on robots operating in factories by aggregating the maintenance knowledge and experience collected from skilled technicians as digital data and using that data to train a generative AI model. This system includes the following main components:

[0556] 1. Digitization of knowledge and experience

[0557] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[0558] 2. Data preprocessing and formatting

[0559] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0560] 3. Training the generative AI model

[0561] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[0562] 4. Providing real-time support

[0563] Users (engineers and new employees) receive support during robot maintenance through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0564] 5. Present and implement response procedures

[0565] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0566] 6. Gather feedback and retrain

[0567] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model for retraining, thereby continuously improving the system's accuracy and efficiency.

[0568] Examples:

[0569] Example 1: Robot arm not moving

[0570] When a user (a new engineer) encounters a problem where the robot arm won't move, they input the details of the problem using a dedicated terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "check the actuator operation," and "check the system log," and provides these to the user via the terminal. The user then takes action according to these steps, and receives feedback to the terminal that the problem has been resolved.

[0571] Example prompt sentence:

[0572] "What is the best procedure to solve the problem of a stuck robot arm?"

[0573] In this way, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[0574] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0575] Step 1:

[0576] The server collects maintenance knowledge and experience from veteran engineers as digital data. Specifically, information such as past trouble cases, problem-solving procedures, and methods for identifying troubles is digitized as text data, images, and videos. The input here is the experience and knowledge of the engineers, which is converted into digital format. The output is maintenance knowledge data stored as a digital archive.

[0577] Step 2:

[0578] The server normalizes and tags the collected digital data. Normalization is the process of standardizing the data format, and tagging is the process of assigning relevant keywords and classification labels. The input is the digital data, which is normalized and tagged. The output is the unified format and tagged data.

[0579] Step 3:

[0580] The server trains the generative AI model using the preprocessed digital data. After the data preprocessing is complete, the AI ​​model is trained using a machine learning algorithm. The input is the preprocessed digital data, and the output is the trained generative AI model.

[0581] Step 4:

[0582] The server integrates the trained generative AI model into the real-time support system. The AI ​​model is incorporated into the entire system so that it can generate maintenance response procedures in real time. The input is the trained generative AI model, and the output is the generative AI model that has become part of the real-time support system.

[0583] Step 5:

[0584] The terminal accepts inquiries from users. The user inputs questions or problems related to maintenance response through the terminal (e.g., smart glasses). The input is the content of the user's inquiry. The output is the content of the inquiry being sent to the server.

[0585] Step 6:

[0586] The server analyzes the inquiry received from the user and generates the optimal response procedure based on the generative AI model. The input is the inquiry from the user, and the output is the response procedure. Specifically, the generative AI model automatically generates the procedure that is most appropriate for the trouble content.

[0587] Step 7:

[0588] The terminal displays the response procedure sent from the server to the user. The input is the response procedure sent from the server, and the output is the specific maintenance response procedure provided to the user. Specifically, the procedure is displayed on the display of the smart glasses.

[0589] Step 8:

[0590] The user actually performs the maintenance work according to the procedure. During the process, the user carries out the necessary checks and repairs based on the procedure presented. The input is the procedure provided from the terminal, and the output is the result of the maintenance work.

[0591] Step 9:

[0592] After the maintenance response is completed, the user inputs feedback on the response results, required time, and response efficiency into the terminal. The input is feedback on the results and efficiency of the maintenance response, and the output is feedback data sent from the terminal to the server.

[0593] Step 10:

[0594] The server uses the collected feedback data to retrain the generative AI model: the input is the feedback data, and the output is a new generative AI model with improved accuracy and efficiency.

[0595] As a specific example, based on the prompt, "Please tell me the best procedure to solve the problem of the robot arm not moving," the user queries the server, and the server generates a procedure for the user to check using smart glasses while carrying out maintenance work.

[0596] 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.

[0597] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system for speeding up and streamlining maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[0598] 1. Digitization of knowledge and experience

[0599] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[0600] 2. Data Preprocessing and Normalization

[0601] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0602] 3. Training the generative AI model

[0603] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0604] 4. Providing real-time support

[0605] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0606] 5. Use of Emotion Engine

[0607] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0608] 6. Emotion-based adjustment of response procedures

[0609] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[0610] 7. Present and implement response procedures

[0611] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[0612] 8. Gather feedback and retrain

[0613] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[0614] Examples:

[0615] Example 1: Screen not displaying issue

[0616] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[0617] Example 2: Communication error problem

[0618] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[0619] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[0620] The processing flow will be explained below.

[0621] Step 1:

[0622] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[0623] Step 2:

[0624] The server normalizes and tags the collected data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0625] Step 3:

[0626] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0627] Step 4:

[0628] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[0629] Step 5:

[0630] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[0631] Step 6:

[0632] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[0633] Step 7:

[0634] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[0635] Step 8:

[0636] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[0637] Step 9:

[0638] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0639] Step 10:

[0640] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[0641] Step 11:

[0642] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[0643] Step 12:

[0644] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[0645] Step 13:

[0646] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[0647] Step 14:

[0648] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and emotion engine, improving the accuracy and efficiency of the system.

[0649] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[0650] Example 2

[0651] 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."

[0652] In on-site maintenance work, new and inexperienced engineers are required to solve problems quickly and efficiently on the spot. However, with conventional systems, it is difficult to share the knowledge and experience gained from veteran engineers in real time, and it is also difficult to respond flexibly while taking into account the user's emotional state. As a result, the quality and efficiency of maintenance work declines.

[0653] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0654] In this invention, the server includes: means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the user inquiries and generating optimal response procedures; means for providing the generated response procedures to the user; means for incorporating an emotion engine into the terminal and recognizing the emotional state of the user from facial expressions and voice; means for adjusting the response procedures based on data from the emotion engine; and means for collecting feedback after maintenance responses and retraining the generative AI model. This enables effective maintenance responses that take the user's emotional state into consideration, thereby improving the quality and efficiency of responses.

[0655] A "technician" is a person who has specialized knowledge and skills related to the maintenance, repair, or operation of a system or equipment.

[0656] "Digital data" refers to maintenance knowledge and experience expressed in an electronic format, and includes text data, images, audio data, and the like.

[0657] "Normalization" refers to the process of converting collected data into a unified format and making it consistent.

[0658] "Tagging" is the process of assigning relevant keywords and classification labels to digital data to make it easier to search and analyze later.

[0659] A "generative AI model" is an artificial intelligence model that can learn from collected and preprocessed data to perform specific tasks.

[0660] "Training" is the process of feeding data and adjusting the parameters of a generative AI model in order to learn useful knowledge and patterns.

[0661] A "real-time support system" is a system that provides immediate support for problems that arise during maintenance.

[0662] "User" refers to the engineers and technicians who use the system to perform maintenance.

[0663] The "emotion engine" is a system that analyzes the user's emotional state from their facial expressions and voice, and determines the appropriate response based on that.

[0664] "Feedback" refers to information about the results and emotional state of a user after a maintenance response.

[0665] "Retraining" is the process of retraining a generative AI model based on newly collected feedback data to improve the model's accuracy and performance.

[0666] The present invention relates to a system for quickly and effectively carrying out maintenance responses, and this system is mainly composed of three elements: a server, a terminal, and a user. How this system functions will be explained in detail below.

[0667] 1. Digitization of knowledge and experience

[0668] The server collects the maintenance knowledge and experience of veteran technicians as digital data. This includes text files, images, and audio data. For example, it collects examples of problems that technicians have solved in the past, along with detailed troubleshooting procedures. This data is stored in a database and can be used for later analysis and retrieval.

[0669] 2. Data Preprocessing and Normalization

[0670] The server converts the collected data into a unified format (normalizes it) and attaches relevant tags to it. Normalization and tagging improve the searchability and usability of the data. For example, adding tags such as "power supply trouble" or "network error" makes it easier to search later.

[0671] 3. Training the generative AI model

[0672] The server uses the preprocessed data to train the generative AI model. During this training process, a large amount of data is fed into the AI ​​model, and by adjusting the weights of the neural network, the AI ​​model learns how to resolve problems. For example, each maintenance procedure and solution method is memorized in the model.

[0673] 4. Providing real-time support

[0674] Users receive support during maintenance work using a dedicated application. The device accepts user input and sends it to a server. The server analyzes the received inquiry and generates optimal response procedures based on a generative AI model. The device then displays the generated response procedures to the user. For example, for a problem such as "the screen is not displaying," specific procedures such as "check the power cable connection" and "restart the system" are provided.

[0675] 5. Use of Emotion Engine

[0676] The device is equipped with an emotion engine that recognizes the user's emotional state from their facial expressions, voice, text, etc. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice and extract information such as whether the user is feeling stressed or relieved.

[0677] 6. Emotion-based adjustment of response procedures

[0678] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, it adds more detailed and careful explanations. In this way, the response procedures can be dynamically adjusted according to the user's emotional state.

[0679] 7. Present and implement response procedures

[0680] The terminal displays the adjusted response procedures to the user. For example, specific steps such as "check the power cable connection" and "measure the voltage" are displayed in a list, and the user follows these to perform the maintenance work.

[0681] 8. Gather feedback and retrain

[0682] After the maintenance response is completed, the user enters feedback into the device about the results, time required, response efficiency, and emotional state. The device then sends the feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine. This continuously improves the accuracy and efficiency of the system.

[0683] Examples:

[0684] Example 1: When a user encounters a problem with a device in a store, such as "the screen is not displaying," they input the details of the problem through the device. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection," "measure the voltage," and "restart the system." The device's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and responds, and the device receives feedback that the problem has been resolved.

[0685] Example 2: When a user encounters a "communication error," they use their device to send the error details to the server. The server uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to allow the user to complete the steps more quickly.

[0686] Example prompt sentence:

[0687] Sample prompt 1: "Please describe a problem a junior engineer encounters while working on a maintenance project in-store. Also, describe how the user's emotional state is reflected when the steps to resolve the problem are displayed. For example, when an error message appears."

[0688] Such a system not only improves the quality and efficiency of maintenance responses, but also enables flexible responses that take into account the user's emotional state.

[0689] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0690] System processing flow

[0691] Step 1: Gathering knowledge and experience

[0692] Input: Knowledge and experience from veteran engineers (text files, images, audio data, etc.)

[0693] The server collects maintenance knowledge and experience from veteran technicians, specifically documents, logs, voice memos, manuals, and other data used by technicians in maintenance operations, as digital data.

[0694] How it works: The server automatically retrieves files of various formats uploaded by technicians and stores them in a database, for example, by importing data from the technician's PC or cloud storage.

[0695] Output: Knowledge and experience in digital data format is stored in a database.

[0696] Step 2: Preprocessing and normalizing data

[0697] Input: Knowledge and experience of digital data formats

[0698] The server normalizes and tags the collected digital data, converting each piece of data into a unified format and assigning relevant keywords and classification labels.

[0699] How it works: The server converts, for example, audio data into text, parses the text data using natural language processing (NLP) to put it into a consistent format, and then assigns tags like "power outage" or "network error."

[0700] Output: A normalized and tagged dataset.

[0701] Step 3: Training the generative AI model

[0702] Input: Normalized and tagged dataset

[0703] The server uses the preprocessed data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods.

[0704] How it works: The server uses a neural network library (such as TensorFlow or PyTorch) to feed the tagged data to a generative AI model, adjusting the model's parameters (weights) to learn the optimal solution procedure.

[0705] Output: A trained generative AI model.

[0706] Step 4: Receiving inquiries from users

[0707] Input: User-entered problem details for maintenance

[0708] The user uses a dedicated application to input details of the problem that is currently being handled by the maintenance team. The terminal accepts this input and sends it to the server.

[0709] How it works: The user enters a specific problem into the application, such as "no screen display," and the device sends that information to the server.

[0710] Output: The query is sent to the server.

[0711] Step 5: Inquiry analysis and response procedure generation

[0712] Input: Inquiry sent by the user

[0713] The server analyzes the received inquiry and generates the optimal response procedure based on the generative AI model.

[0714] How it works: The server uses natural language processing to analyze the inquiry and generate specific response procedures such as "check the power cable connection" and "restart the system."

[0715] Output: The optimal response procedure is generated.

[0716] Step 6: View response procedures

[0717] Input: Generated response procedure

[0718] The terminal displays the generated response procedure to the user.

[0719] What it does: The device application will display instructions in text format, such as "Check the power cable. Then measure the voltage."

[0720] Output: Instructions for action are displayed to the user.

[0721] Step 7: Use the Emotion Engine

[0722] Input: User's facial expression and voice data

[0723] The terminal uses an emotion engine to recognize the user's emotional state from their facial expressions and voice.

[0724] How it works: The device analyzes images and audio captured by the camera and microphone to determine the user's emotions. For example, if the user is frowning, it will determine that the user is in a state of stress.

[0725] Output: User's emotional state data.

[0726] Step 8: Adjust response procedures

[0727] Input: User emotional state data

[0728] The server adjusts the response procedures generated by the generative AI model based on the emotional data from the emotion engine.

[0729] What it does: The server adds reassuring statements like "Don't worry" to the steps if the user is feeling anxious, or simplifies the steps if the user is feeling impatient.

[0730] Output: Coordinated response procedures.

[0731] Step 9: View the adjusted steps

[0732] Input: Coordinated response procedures

[0733] The terminal displays the adjusted procedure to the user.

[0734] What it does: The adjusted steps are displayed on the device, and the user can refer to them to perform the maintenance action. For example, the message "Don't worry, this step will be over soon."

[0735] Output: Adjusted response instructions displayed to the user.

[0736] Step 10: Gather feedback

[0737] Inputs: Maintenance response results, response time, response efficiency, emotional state

[0738] After the maintenance is completed, the user inputs feedback into the terminal.

[0739] Action: The user enters feedback such as "Problem solved", "Took 15 minutes", and "I was anxious during the process".

[0740] Output: Feedback data is sent from the device to the server.

[0741] Step 11: Retraining the generative AI model and emotion engine

[0742] Input: Collected feedback data

[0743] The server retrains the generative AI model and emotion engine based on newly collected feedback data.

[0744] How it works: The server uses the feedback data to readjust the parameters of the generative AI model to improve the accuracy and efficiency of the system, and also retrains the emotion engine to more accurately judge user emotions.

[0745] Output: Retrained generative AI model and emotion engine.

[0746] (Application example 2)

[0747] 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."

[0748] Existing maintenance support systems rely on the knowledge and experience of the technician, making it difficult for new or inexperienced technicians to provide effective maintenance support. Furthermore, support that does not take into account the user's emotional state can cause stress and anxiety. Another issue is the lack of real-time support, which means it takes a long time to resolve problems.

[0749] 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.

[0750] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance responses and retraining the generative AI model; means for adjusting the generated response procedures using an emotion engine that recognizes the user's emotions; and means for collecting and analyzing the user's emotions using a camera and microphone of a smart device, thereby enabling flexible and effective real-time support according to the user's emotional state.

[0751] "Collecting knowledge and experience" is the process of collecting knowledge and experience regarding maintenance responses from veteran technicians as digital data.

[0752] "Normalization and tagging" is the process of converting collected digital data into a unified format and assigning relevant keywords and classification labels.

[0753] "Training a generative AI model" is the process of using normalized and tagged data to teach a generative AI model how to respond efficiently.

[0754] "Real-time support system integration" is the process of building a system that uses trained generative AI models to provide support to users in real time.

[0755] "Accepting inquiries from users" is a means by which users can send questions about maintenance responses or trouble reports to the system.

[0756] "Analysis of inquiries and generation of response procedures" is a process of analyzing an inquiry from a user and generating an optimal response procedure based on that analysis.

[0757] An "emotion engine" is an engine that recognizes and analyzes a user's emotional state from facial expressions, voice, text, etc.

[0758] "Use of smart devices" refers to the use of devices such as smartphones and tablets to collect the user's emotional state through cameras and microphones.

[0759] "Feedback collection and retraining" refers to the process of collecting feedback from users after maintenance is performed and retraining the generative AI model based on that feedback.

[0760] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system that speeds up and streamlines maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[0761] 1. Digitization of knowledge and experience

[0762] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[0763] 2. Data Preprocessing and Normalization

[0764] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0765] 3. Training the generative AI model

[0766] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0767] 4. Providing real-time support

[0768] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0769] 5. Use of Emotion Engine

[0770] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0771] 6. Emotion-based adjustment of response procedures

[0772] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[0773] 7. Present and implement response procedures

[0774] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[0775] 8. Gather feedback and retrain

[0776] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[0777] Examples:

[0778] Example 1: Screen not displaying issue

[0779] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[0780] Example 2: Communication error problem

[0781] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[0782] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[0783] Example prompt sentence:

[0784] If the user types "screen not showing":

[0785] User is feeling anxious. Provide detailed and comforting support message. Base message: "Please check the connection of the power cable of the screen."

[0786] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0787] Step 1:

[0788] The terminal accepts the contents of a maintenance inquiry from the user as input. The user inputs the inquiry into the terminal, and the terminal transmits the inquiry contents to the server as text data.

[0789] Step 2:

[0790] The server analyzes the inquiry received from the device. The server uses a generative AI model to analyze the inquiry and generate appropriate response procedures. The server uses the text data of the inquiry as input and provides the generated text of the response procedures as output.

[0791] Step 3:

[0792] The terminal provides the user with the response procedure sent from the server, and receives the generated response procedure text from the server and displays it to the user in an appropriate format.

[0793] Step 4:

[0794] The device uses an emotion engine to collect the user's emotional state as input. It uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to obtain emotion data. It uses camera images and audio data as input and obtains emotional state data as output.

[0795] Step 5:

[0796] The server adjusts the response procedure created by the generative AI model based on the acquired emotional state data. It receives the emotional state data as input and inputs the prompt sentence into the generative AI model to create a newly adjusted response procedure. As an output, it obtains the text of the response procedure adjusted based on the emotion.

[0797] Step 6:

[0798] The terminal displays the adjusted response procedure to the user. The terminal receives the text of the adjusted response procedure from the server and displays it to the user.

[0799] Step 7:

[0800] The user performs maintenance and inputs the results as feedback to the terminal. After the maintenance is completed, the results, required time, emotional state, etc. are input to the terminal, and the terminal sends the contents to the server. The text data of the feedback content is used as input, and feedback data is provided to be sent to the server as output.

[0801] Step 8:

[0802] The server retrains the generative AI model using the received feedback data. It receives the feedback data as input, adjusts the weights of the generative AI model based on the feedback data, and retrains the model. It provides an updated generative AI model as output.

[0803] 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.

[0804] 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.

[0805] 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.

[0806] [Third embodiment]

[0807] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0808] 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.

[0809] 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).

[0810] 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.

[0811] 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.

[0812] 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).

[0813] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[0814] 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.

[0815] 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.

[0816] 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.

[0817] 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.

[0818] 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."

[0819] This invention is a system for speeding up and streamlining maintenance of storefront equipment, utilizing a generative AI model to aggregate the knowledge and experience of veterans, enabling all staff to work as experts. The system includes the following main components:

[0820] 1. Digitization of knowledge and experience

[0821] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[0822] 2. Data preprocessing and formatting

[0823] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0824] 3. Training the generative AI model

[0825] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[0826] 4. Providing real-time support

[0827] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0828] 5. Present and implement response procedures

[0829] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0830] 6. Gather feedback and retrain

[0831] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model and retrains it, continuously improving the system's accuracy and efficiency.

[0832] Examples:

[0833] Example 1: Screen not displaying issue

[0834] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The user then takes action according to the instructions, and the terminal receives feedback that the problem has been resolved.

[0835] Example 2: Communication error problem

[0836] When a user encounters a communication error, they use their device to send the error details to the server. The server uses a generative AI model to generate steps such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The user then follows the steps and responds, receiving feedback on the final result.

[0837] As described above, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[0838] The processing flow will be explained below.

[0839] Step 1:

[0840] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[0841] Step 2:

[0842] The server normalizes and tags the collected data, standardizing its format and adding relevant keywords and labels, a process that makes the data more searchable and useful.

[0843] Step 3:

[0844] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0845] Step 4:

[0846] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[0847] Step 5:

[0848] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[0849] Step 6:

[0850] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[0851] Step 7:

[0852] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[0853] Step 8:

[0854] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[0855] Step 9:

[0856] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[0857] Step 10:

[0858] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[0859] Step 11:

[0860] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and improve the accuracy and efficiency of the system.

[0861] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[0862] Example 1

[0863] 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."

[0864] In conventional maintenance response systems, the experience and knowledge of technicians was highly dependent on the individual, making it difficult for new technicians to respond quickly and accurately. Furthermore, there was a lack of a way to efficiently share the know-how of veteran technicians, making it an issue to streamline troubleshooting.

[0865] 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.

[0866] In this invention, the server includes a means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data, a means for normalizing and tagging the digital data, and a means for training a generative AI model using the normalized and tagged data. This allows maintenance response knowledge to be digitized and standardized, and efficiently shared with new technicians through the AI ​​model. It also supports real-time troubleshooting, improving the quality and efficiency of maintenance response.

[0867] A "veteran technician" is a maintenance specialist with many years of experience and advanced technical knowledge.

[0868] "Knowledge and experience in maintenance" refers to troubleshooting and problem-solving procedures and related information obtained during maintenance work.

[0869] "Digital data" refers to information such as text, images, and video that has been converted into a form that a computer can process.

[0870] "Normalization" is the process of putting data into a consistent format, which is done to ensure data integrity and consistency.

[0871] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[0872] A "generative AI model" is an artificial intelligence model that learns problem-solving procedures and efficient response methods based on collected data and automatically generates them.

[0873] A "real-time support system" is a system that allows new engineers and others to receive immediate support while performing maintenance work.

[0874] "Means for accepting inquiries" refers to an interface or device for receiving questions or problems from users.

[0875] "Means for generating response procedures" refers to methods and techniques for creating optimal solution procedures based on the content of inquiries received from users.

[0876] "Means for collecting feedback" refers to a method or system for collecting evaluations and result information from users after maintenance has been performed.

[0877] "Retraining" is the process of re-learning an AI model using new data collected to improve its accuracy and efficiency.

[0878] "Data preprocessing and formatting" refers to a series of procedures for converting collected data into a format suitable for analysis or training.

[0879] This invention is a system for speeding up and streamlining maintenance responses for storefront equipment. It utilizes generative AI models to aggregate the knowledge and experience of veterans, providing a mechanism for all staff to respond as experts. The system includes the following main components:

[0880] 1. Digitization of knowledge and experience

[0881] The server collects maintenance knowledge and experience from veteran technicians and stores it as digital data. Specifically, it includes information such as past trouble cases recorded by technicians, problem-solving procedures, and methods for identifying troubles. This digital data is stored in a database and used for later processing. The collected data is stored in text, image, and video formats.

[0882] 2. Data preprocessing and formatting

[0883] The server normalizes the collected digital data and compiles it into a unified format. For example, it converts trouble cases recorded in different formats into a unified text format. It also tags the data with relevant keywords and classification labels. This improves data searchability and enables more efficient use.

[0884] 3. Training the generative AI model

[0885] The server uses preprocessed digital data to train a generative AI model. Large amounts of historical data are used, particularly to learn troubleshooting procedures and efficient response methods. This AI model is then trained using machine learning algorithms, giving it the ability to flexibly respond to new problems.

[0886] 4. Providing real-time support

[0887] Users (engineers and new employees) receive support for maintenance responses through a dedicated application. When the user inputs the problem into the device, the device sends the input information to the server. The server analyzes the received information and generates the optimal response procedure using a generative AI model. The generated procedure is then sent to the device, which displays it to the user.

[0888] 5. Present and implement response procedures

[0889] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0890] 6. Gather feedback and retrain

[0891] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, and response efficiency. The device then sends this feedback to the server, which processes it and reflects it in the generated AI model. The AI ​​model is retrained based on the feedback data, thereby continuously improving the accuracy and efficiency of the entire system.

[0892] Examples:

[0893] Example 1: Screen not displaying issue

[0894] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input "screen not displaying" into the terminal. The server analyzes this information and uses a generative AI model to suggest next steps:

[0895] 1. Check the power cable connection.

[0896] 2. Check that the voltage is correct.

[0897] 3. Reboot your system.

[0898] The user responds accordingly and provides feedback to the terminal that the problem has been solved.

[0899] Example 2: Communication error problem

[0900] When a user encounters a communication error, they input "communication error" into their device. The server uses a generative AI model to suggest the following steps:

[0901] 1. Check the cable connections.

[0902] 2. Restart your router.

[0903] 3. Check the server settings.

[0904] The user follows the procedure and responds by providing feedback on the final result.

[0905] As described above, this invention improves the quality and efficiency of maintenance response by consolidating the knowledge and experience of veterans and providing support in real time.

[0906] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0907] Step 1:

[0908] The server collects maintenance knowledge and experience from veteran engineers. Specifically, it collects data such as trouble cases recorded by engineers in the past, problem-solving procedures, and identification methods, and stores it in a database. The input is notes and reports on the engineers' experience and knowledge, and the output is digital data in a unified format.

[0909] Step 2:

[0910] The server normalizes and tags the collected digital data. The input is the digital data collected in step 1, and normalization and tagging improve the consistency and searchability of the data. The output is normalized and tagged data. Specifically, data in different formats is converted into a unified format and keywords and classification labels are assigned.

[0911] Step 3:

[0912] The server trains the generative AI model using the preprocessed digital data. The input is the data processed in step 2, and the machine learning algorithm is used to train the AI ​​model. The output is the trained generative AI model, which can generate troubleshooting procedures.

[0913] Step 4:

[0914] The user inputs the problem into the terminal. The specific input is a symptom of the problem, such as "the screen is not displayed" or "communication error." The terminal receives this input and sends it to the server. The input is the user's problem, and the output is the data to be sent.

[0915] Step 5:

[0916] The server analyzes the problem sent by the user and generates the optimal response procedure. The input is the problem data sent in step 4, and the response procedure is automatically generated using a generative AI model. The output is the generated response procedure. Specifically, the problem data is analyzed, and procedures such as "checking the power cable connection" and "measuring voltage" are created using the AI ​​model.

[0917] Step 6:

[0918] The server sends the generated response procedure to the terminal. The input is the response procedure generated in step 5, and the output is the data to be sent to the terminal. The specific operation is to transfer the generated procedure to the terminal.

[0919] Step 7:

[0920] The terminal displays the response procedures it has received to the user. The input is the response procedure data received in step 6, and the output is the display to the user. Specifically, the response procedures are displayed in list format on the terminal screen.

[0921] Step 8:

[0922] The user performs maintenance actions by following the procedures displayed on the terminal. The input is the procedures displayed in step 7, and the output is the results of performing the maintenance actions. The specific actions are the user performing physical operations based on the procedures.

[0923] Step 9:

[0924] The user inputs the results of the maintenance work into the terminal. Specific inputs include "the problem has been resolved" or "an additional problem has occurred." The terminal sends this feedback to the server. The input is the user's feedback, and the output is the data to send it.

[0925] Step 10:

[0926] The server processes the feedback and retrains the generative AI model. The input is the feedback data sent in step 9, and the output is the retrained generative AI model. The specific operation is to analyze the feedback data and update the AI ​​model taking into account the new data.

[0927] (Application example 1)

[0928] 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."

[0929] On-site maintenance is particularly difficult for new employees and inexperienced technicians, and requires a fast and accurate response. In addition, the robots used in factories have complex structures, so a high level of specialized knowledge is required to respond when a problem occurs. For this reason, a system is needed that allows new technicians to respond quickly and at the same level as experienced technicians.

[0930] 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.

[0931] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran engineers as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance and retraining the generative AI model; means for providing information to a dedicated device for performing maintenance on robots operating in factories; and means for displaying the maintenance response procedures to the user using the dedicated device. This enables even novice engineers to perform robot maintenance quickly and accurately.

[0932] A "veteran engineer" is an engineer who has advanced specialized knowledge and practical skills based on many years of experience.

[0933] "Maintenance response" refers to work such as inspection, repair, and adjustment carried out to maintain the normal operation of equipment and systems.

[0934] "Digital data" refers to data handled by digital devices such as computers, and includes text, images, videos, etc.

[0935] "Normalization" is the process of converting data into a unified format.

[0936] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[0937] A "generative AI model" is an artificial intelligence model that is trained using machine learning algorithms based on collected data.

[0938] A "real-time support system" is a system designed to allow users to receive support in real time.

[0939] "User inquiries" refer to the act of a user reporting a question or problem regarding maintenance support to the system.

[0940] "Specialized equipment" refers to devices or equipment specifically used for this invention.

[0941] "Means provided to the user" refers to the method by which the system presents the user with the most appropriate response procedures and information.

[0942] "Feedback after maintenance" refers to evaluations and opinions provided by users after maintenance work has been completed.

[0943] "Retraining" is the process of retraining a previously trained AI model with new data to improve the model's accuracy or performance.

[0944] "Robots operating in factories" refers to automated robots used on factory production lines and in workshops.

[0945] "Maintenance response procedures" refer to the specific procedures and steps for performing maintenance work on equipment or systems.

[0946] This invention is a system that efficiently performs maintenance on robots operating in factories by aggregating the maintenance knowledge and experience collected from skilled technicians as digital data and using that data to train a generative AI model. This system includes the following main components:

[0947] 1. Digitization of knowledge and experience

[0948] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[0949] 2. Data preprocessing and formatting

[0950] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0951] 3. Training the generative AI model

[0952] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[0953] 4. Providing real-time support

[0954] Users (engineers and new employees) receive support during robot maintenance through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0955] 5. Present and implement response procedures

[0956] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[0957] 6. Gather feedback and retrain

[0958] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model for retraining, thereby continuously improving the system's accuracy and efficiency.

[0959] Examples:

[0960] Example 1: Robot arm not moving

[0961] When a user (a new engineer) encounters a problem where the robot arm won't move, they input the details of the problem using a dedicated terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "check the actuator operation," and "check the system log," and provides these to the user via the terminal. The user then takes action according to these steps, and receives feedback to the terminal that the problem has been resolved.

[0962] Example prompt sentence:

[0963] "What is the best procedure to solve the problem of a stuck robot arm?"

[0964] In this way, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[0965] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0966] Step 1:

[0967] The server collects maintenance knowledge and experience from veteran engineers as digital data. Specifically, information such as past trouble cases, problem-solving procedures, and methods for identifying troubles is digitized as text data, images, and videos. The input here is the experience and knowledge of the engineers, which is converted into digital format. The output is maintenance knowledge data stored as a digital archive.

[0968] Step 2:

[0969] The server normalizes and tags the collected digital data. Normalization is the process of standardizing the data format, and tagging is the process of assigning relevant keywords and classification labels. The input is the digital data, which is normalized and tagged. The output is the unified format and tagged data.

[0970] Step 3:

[0971] The server trains the generative AI model using the preprocessed digital data. After the data preprocessing is complete, the AI ​​model is trained using a machine learning algorithm. The input is the preprocessed digital data, and the output is the trained generative AI model.

[0972] Step 4:

[0973] The server integrates the trained generative AI model into the real-time support system. The AI ​​model is incorporated into the entire system so that it can generate maintenance response procedures in real time. The input is the trained generative AI model, and the output is the generative AI model that has become part of the real-time support system.

[0974] Step 5:

[0975] The terminal accepts inquiries from users. The user inputs questions or problems related to maintenance response through the terminal (e.g., smart glasses). The input is the content of the user's inquiry. The output is the content of the inquiry being sent to the server.

[0976] Step 6:

[0977] The server analyzes the inquiry received from the user and generates the optimal response procedure based on the generative AI model. The input is the inquiry from the user, and the output is the response procedure. Specifically, the generative AI model automatically generates the procedure that is most appropriate for the trouble content.

[0978] Step 7:

[0979] The terminal displays the response procedure sent from the server to the user. The input is the response procedure sent from the server, and the output is the specific maintenance response procedure provided to the user. Specifically, the procedure is displayed on the display of the smart glasses.

[0980] Step 8:

[0981] The user actually performs the maintenance work according to the procedure. During the process, the user carries out the necessary checks and repairs based on the procedure presented. The input is the procedure provided from the terminal, and the output is the result of the maintenance work.

[0982] Step 9:

[0983] After the maintenance response is completed, the user inputs feedback on the response results, required time, and response efficiency into the terminal. The input is feedback on the results and efficiency of the maintenance response, and the output is feedback data sent from the terminal to the server.

[0984] Step 10:

[0985] The server uses the collected feedback data to retrain the generative AI model: the input is the feedback data, and the output is a new generative AI model with improved accuracy and efficiency.

[0986] As a specific example, based on the prompt, "Please tell me the best procedure to solve the problem of the robot arm not moving," the user queries the server, and the server generates a procedure for the user to check using smart glasses while carrying out maintenance work.

[0987] 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.

[0988] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system for speeding up and streamlining maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[0989] 1. Digitization of knowledge and experience

[0990] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[0991] 2. Data Preprocessing and Normalization

[0992] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[0993] 3. Training the generative AI model

[0994] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[0995] 4. Providing real-time support

[0996] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[0997] 5. Use of Emotion Engine

[0998] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[0999] 6. Emotion-based adjustment of response procedures

[1000] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[1001] 7. Present and implement response procedures

[1002] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[1003] 8. Gather feedback and retrain

[1004] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[1005] Examples:

[1006] Example 1: Screen not displaying issue

[1007] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[1008] Example 2: Communication error problem

[1009] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[1010] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[1011] The processing flow will be explained below.

[1012] Step 1:

[1013] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[1014] Step 2:

[1015] The server normalizes and tags the collected data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1016] Step 3:

[1017] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[1018] Step 4:

[1019] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[1020] Step 5:

[1021] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[1022] Step 6:

[1023] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[1024] Step 7:

[1025] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[1026] Step 8:

[1027] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[1028] Step 9:

[1029] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[1030] Step 10:

[1031] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[1032] Step 11:

[1033] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[1034] Step 12:

[1035] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[1036] Step 13:

[1037] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[1038] Step 14:

[1039] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and emotion engine, improving the accuracy and efficiency of the system.

[1040] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[1041] Example 2

[1042] 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."

[1043] In on-site maintenance work, new and inexperienced engineers are required to solve problems quickly and efficiently on the spot. However, with conventional systems, it is difficult to share the knowledge and experience gained from veteran engineers in real time, and it is also difficult to respond flexibly while taking into account the user's emotional state. As a result, the quality and efficiency of maintenance work declines.

[1044] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1045] In this invention, the server includes: means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the user inquiries and generating optimal response procedures; means for providing the generated response procedures to the user; means for incorporating an emotion engine into the terminal and recognizing the emotional state of the user from facial expressions and voice; means for adjusting the response procedures based on data from the emotion engine; and means for collecting feedback after maintenance responses and retraining the generative AI model. This enables effective maintenance responses that take the user's emotional state into consideration, thereby improving the quality and efficiency of responses.

[1046] A "technician" is a person who has specialized knowledge and skills related to the maintenance, repair, or operation of a system or equipment.

[1047] "Digital data" refers to maintenance knowledge and experience expressed in an electronic format, and includes text data, images, audio data, and the like.

[1048] "Normalization" refers to the process of converting collected data into a unified format and making it consistent.

[1049] "Tagging" is the process of assigning relevant keywords and classification labels to digital data to make it easier to search and analyze later.

[1050] A "generative AI model" is an artificial intelligence model that can learn from collected and preprocessed data to perform specific tasks.

[1051] "Training" is the process of feeding data and adjusting the parameters of a generative AI model in order to learn useful knowledge and patterns.

[1052] A "real-time support system" is a system that provides immediate support for problems that arise during maintenance.

[1053] "User" refers to the engineers and technicians who use the system to perform maintenance.

[1054] The "emotion engine" is a system that analyzes the user's emotional state from their facial expressions and voice, and determines the appropriate response based on that.

[1055] "Feedback" refers to information about the results and emotional state of a user after a maintenance response.

[1056] "Retraining" is the process of retraining a generative AI model based on newly collected feedback data to improve the model's accuracy and performance.

[1057] The present invention relates to a system for quickly and effectively carrying out maintenance responses, and this system is mainly composed of three elements: a server, a terminal, and a user. How this system functions will be explained in detail below.

[1058] 1. Digitization of knowledge and experience

[1059] The server collects the maintenance knowledge and experience of veteran technicians as digital data. This includes text files, images, and audio data. For example, it collects examples of problems that technicians have solved in the past, along with detailed troubleshooting procedures. This data is stored in a database and can be used for later analysis and retrieval.

[1060] 2. Data Preprocessing and Normalization

[1061] The server converts the collected data into a unified format (normalizes it) and attaches relevant tags to it. Normalization and tagging improve the searchability and usability of the data. For example, adding tags such as "power supply trouble" or "network error" makes it easier to search later.

[1062] 3. Training the generative AI model

[1063] The server uses the preprocessed data to train the generative AI model. During this training process, a large amount of data is fed into the AI ​​model, and by adjusting the weights of the neural network, the AI ​​model learns how to resolve problems. For example, each maintenance procedure and solution method is memorized in the model.

[1064] 4. Providing real-time support

[1065] Users receive support during maintenance work using a dedicated application. The device accepts user input and sends it to a server. The server analyzes the received inquiry and generates optimal response procedures based on a generative AI model. The device then displays the generated response procedures to the user. For example, for a problem such as "the screen is not displaying," specific procedures such as "check the power cable connection" and "restart the system" are provided.

[1066] 5. Use of Emotion Engine

[1067] The device is equipped with an emotion engine that recognizes the user's emotional state from their facial expressions, voice, text, etc. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice and extract information such as whether the user is feeling stressed or relieved.

[1068] 6. Emotion-based adjustment of response procedures

[1069] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, it adds more detailed and careful explanations. In this way, the response procedures can be dynamically adjusted according to the user's emotional state.

[1070] 7. Present and implement response procedures

[1071] The terminal displays the adjusted response procedures to the user. For example, specific steps such as "check the power cable connection" and "measure the voltage" are displayed in a list, and the user follows these to perform the maintenance work.

[1072] 8. Gather feedback and retrain

[1073] After the maintenance response is completed, the user enters feedback into the device about the results, time required, response efficiency, and emotional state. The device then sends the feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine. This continuously improves the accuracy and efficiency of the system.

[1074] Examples:

[1075] Example 1: When a user encounters a problem with a device in a store, such as "the screen is not displaying," they input the details of the problem through the device. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection," "measure the voltage," and "restart the system." The device's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and responds, and the device receives feedback that the problem has been resolved.

[1076] Example 2: When a user encounters a "communication error," they use their device to send the error details to the server. The server uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to allow the user to complete the steps more quickly.

[1077] Example prompt sentence:

[1078] Sample prompt 1: "Please describe a problem a junior engineer encounters while working on a maintenance project in-store. Also, describe how the user's emotional state is reflected when the steps to resolve the problem are displayed. For example, when an error message appears."

[1079] Such a system not only improves the quality and efficiency of maintenance responses, but also enables flexible responses that take into account the user's emotional state.

[1080] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1081] System processing flow

[1082] Step 1: Gathering knowledge and experience

[1083] Input: Knowledge and experience from veteran engineers (text files, images, audio data, etc.)

[1084] The server collects maintenance knowledge and experience from veteran technicians, specifically documents, logs, voice memos, manuals, and other data used by technicians in maintenance operations, as digital data.

[1085] How it works: The server automatically retrieves files of various formats uploaded by technicians and stores them in a database, for example, by importing data from the technician's PC or cloud storage.

[1086] Output: Knowledge and experience in digital data format is stored in a database.

[1087] Step 2: Preprocessing and normalizing data

[1088] Input: Knowledge and experience of digital data formats

[1089] The server normalizes and tags the collected digital data, converting each piece of data into a unified format and assigning relevant keywords and classification labels.

[1090] How it works: The server converts, for example, audio data into text, parses the text data using natural language processing (NLP) to put it into a consistent format, and then assigns tags like "power outage" or "network error."

[1091] Output: A normalized and tagged dataset.

[1092] Step 3: Training the generative AI model

[1093] Input: Normalized and tagged dataset

[1094] The server uses the preprocessed data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods.

[1095] How it works: The server uses a neural network library (such as TensorFlow or PyTorch) to feed the tagged data to a generative AI model, adjusting the model's parameters (weights) to learn the optimal solution procedure.

[1096] Output: A trained generative AI model.

[1097] Step 4: Receiving inquiries from users

[1098] Input: User-entered problem details for maintenance

[1099] The user uses a dedicated application to input details of the problem that is currently being handled by the maintenance team. The terminal accepts this input and sends it to the server.

[1100] How it works: The user enters a specific problem into the application, such as "no screen display," and the device sends that information to the server.

[1101] Output: The query is sent to the server.

[1102] Step 5: Inquiry analysis and response procedure generation

[1103] Input: Inquiry sent by the user

[1104] The server analyzes the received inquiry and generates the optimal response procedure based on the generative AI model.

[1105] How it works: The server uses natural language processing to analyze the inquiry and generate specific response procedures such as "check the power cable connection" and "restart the system."

[1106] Output: The optimal response procedure is generated.

[1107] Step 6: View response procedures

[1108] Input: Generated response procedure

[1109] The terminal displays the generated response procedure to the user.

[1110] What it does: The device application will display instructions in text format, such as "Check the power cable. Then measure the voltage."

[1111] Output: Instructions for action are displayed to the user.

[1112] Step 7: Use the Emotion Engine

[1113] Input: User's facial expression and voice data

[1114] The terminal uses an emotion engine to recognize the user's emotional state from their facial expressions and voice.

[1115] How it works: The device analyzes images and audio captured by the camera and microphone to determine the user's emotions. For example, if the user is frowning, it will determine that the user is in a state of stress.

[1116] Output: User's emotional state data.

[1117] Step 8: Adjust response procedures

[1118] Input: User emotional state data

[1119] The server adjusts the response procedures generated by the generative AI model based on the emotional data from the emotion engine.

[1120] What it does: The server adds reassuring statements like "Don't worry" to the steps if the user is feeling anxious, or simplifies the steps if the user is feeling impatient.

[1121] Output: Coordinated response procedures.

[1122] Step 9: View the adjusted steps

[1123] Input: Coordinated response procedures

[1124] The terminal displays the adjusted procedure to the user.

[1125] What it does: The adjusted steps are displayed on the device, and the user can refer to them to perform the maintenance action. For example, the message "Don't worry, this step will be over soon."

[1126] Output: Adjusted response instructions displayed to the user.

[1127] Step 10: Gather feedback

[1128] Inputs: Maintenance response results, response time, response efficiency, emotional state

[1129] After the maintenance is completed, the user inputs feedback into the terminal.

[1130] Action: The user enters feedback such as "Problem solved", "Took 15 minutes", and "I was anxious during the process".

[1131] Output: Feedback data is sent from the device to the server.

[1132] Step 11: Retraining the generative AI model and emotion engine

[1133] Input: Collected feedback data

[1134] The server retrains the generative AI model and emotion engine based on newly collected feedback data.

[1135] How it works: The server uses the feedback data to readjust the parameters of the generative AI model to improve the accuracy and efficiency of the system, and also retrains the emotion engine to more accurately judge user emotions.

[1136] Output: Retrained generative AI model and emotion engine.

[1137] (Application example 2)

[1138] 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."

[1139] Existing maintenance support systems rely on the knowledge and experience of the technician, making it difficult for new or inexperienced technicians to provide effective maintenance support. Furthermore, support that does not take into account the user's emotional state can cause stress and anxiety. Another issue is the lack of real-time support, which means it takes a long time to resolve problems.

[1140] 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.

[1141] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance responses and retraining the generative AI model; means for adjusting the generated response procedures using an emotion engine that recognizes the user's emotions; and means for collecting and analyzing the user's emotions using a camera and microphone of a smart device, thereby enabling flexible and effective real-time support according to the user's emotional state.

[1142] "Collecting knowledge and experience" is the process of collecting knowledge and experience regarding maintenance responses from veteran technicians as digital data.

[1143] "Normalization and tagging" is the process of converting collected digital data into a unified format and assigning relevant keywords and classification labels.

[1144] "Training a generative AI model" is the process of using normalized and tagged data to teach a generative AI model how to respond efficiently.

[1145] "Real-time support system integration" is the process of building a system that uses trained generative AI models to provide support to users in real time.

[1146] "Accepting inquiries from users" is a means by which users can send questions about maintenance responses or trouble reports to the system.

[1147] "Analysis of inquiries and generation of response procedures" is a process of analyzing an inquiry from a user and generating an optimal response procedure based on that analysis.

[1148] An "emotion engine" is an engine that recognizes and analyzes a user's emotional state from facial expressions, voice, text, etc.

[1149] "Use of smart devices" refers to the use of devices such as smartphones and tablets to collect the user's emotional state through cameras and microphones.

[1150] "Feedback collection and retraining" refers to the process of collecting feedback from users after maintenance is performed and retraining the generative AI model based on that feedback.

[1151] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system that speeds up and streamlines maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[1152] 1. Digitization of knowledge and experience

[1153] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[1154] 2. Data Preprocessing and Normalization

[1155] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1156] 3. Training the generative AI model

[1157] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[1158] 4. Providing real-time support

[1159] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[1160] 5. Use of Emotion Engine

[1161] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[1162] 6. Emotion-based adjustment of response procedures

[1163] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[1164] 7. Present and implement response procedures

[1165] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[1166] 8. Gather feedback and retrain

[1167] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[1168] Examples:

[1169] Example 1: Screen not displaying issue

[1170] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[1171] Example 2: Communication error problem

[1172] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[1173] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[1174] Example prompt sentence:

[1175] If the user types "screen not showing":

[1176] User is feeling anxious. Provide detailed and comforting support message. Base message: "Please check the connection of the power cable of the screen."

[1177] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1178] Step 1:

[1179] The terminal accepts the contents of a maintenance inquiry from the user as input. The user inputs the inquiry into the terminal, and the terminal transmits the inquiry contents to the server as text data.

[1180] Step 2:

[1181] The server analyzes the inquiry received from the device. The server uses a generative AI model to analyze the inquiry and generate appropriate response procedures. The server uses the text data of the inquiry as input and provides the generated text of the response procedures as output.

[1182] Step 3:

[1183] The terminal provides the user with the response procedure sent from the server, and receives the generated response procedure text from the server and displays it to the user in an appropriate format.

[1184] Step 4:

[1185] The device uses an emotion engine to collect the user's emotional state as input. It uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to obtain emotion data. It uses camera images and audio data as input and obtains emotional state data as output.

[1186] Step 5:

[1187] The server adjusts the response procedure created by the generative AI model based on the acquired emotional state data. It receives the emotional state data as input and inputs the prompt sentence into the generative AI model to create a newly adjusted response procedure. As an output, it obtains the text of the response procedure adjusted based on the emotion.

[1188] Step 6:

[1189] The terminal displays the adjusted response procedure to the user. The terminal receives the text of the adjusted response procedure from the server and displays it to the user.

[1190] Step 7:

[1191] The user performs maintenance and inputs the results as feedback to the terminal. After the maintenance is completed, the results, required time, emotional state, etc. are input to the terminal, and the terminal sends the contents to the server. The text data of the feedback content is used as input, and feedback data is provided to be sent to the server as output.

[1192] Step 8:

[1193] The server retrains the generative AI model using the received feedback data. It receives the feedback data as input, adjusts the weights of the generative AI model based on the feedback data, and retrains the model. It provides an updated generative AI model as output.

[1194] 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.

[1195] 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.

[1196] 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.

[1197] [Fourth embodiment]

[1198] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1199] 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.

[1200] 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).

[1201] 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.

[1202] 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.

[1203] 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).

[1204] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for 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.

[1205] 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.

[1206] 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.

[1207] 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.

[1208] 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.

[1209] 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.

[1210] 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."

[1211] This invention is a system for speeding up and streamlining maintenance of storefront equipment, utilizing a generative AI model to aggregate the knowledge and experience of veterans, enabling all staff to work as experts. The system includes the following main components:

[1212] 1. Digitization of knowledge and experience

[1213] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[1214] 2. Data preprocessing and formatting

[1215] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1216] 3. Training the generative AI model

[1217] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[1218] 4. Providing real-time support

[1219] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[1220] 5. Present and implement response procedures

[1221] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[1222] 6. Gather feedback and retrain

[1223] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model and retrains it, continuously improving the system's accuracy and efficiency.

[1224] Examples:

[1225] Example 1: Screen not displaying issue

[1226] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The user then takes action according to the instructions, and the terminal receives feedback that the problem has been resolved.

[1227] Example 2: Communication error problem

[1228] When a user encounters a communication error, they use their device to send the error details to the server. The server uses a generative AI model to generate steps such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The user then follows the steps and responds, receiving feedback on the final result.

[1229] As described above, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[1230] The processing flow will be explained below.

[1231] Step 1:

[1232] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[1233] Step 2:

[1234] The server normalizes and tags the collected data, standardizing its format and adding relevant keywords and labels, a process that makes the data more searchable and useful.

[1235] Step 3:

[1236] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[1237] Step 4:

[1238] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[1239] Step 5:

[1240] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[1241] Step 6:

[1242] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[1243] Step 7:

[1244] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[1245] Step 8:

[1246] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[1247] Step 9:

[1248] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[1249] Step 10:

[1250] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[1251] Step 11:

[1252] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and improve the accuracy and efficiency of the system.

[1253] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[1254] Example 1

[1255] 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."

[1256] In conventional maintenance response systems, the experience and knowledge of technicians was highly dependent on the individual, making it difficult for new technicians to respond quickly and accurately. Furthermore, there was a lack of a way to efficiently share the know-how of veteran technicians, making it an issue to streamline troubleshooting.

[1257] 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.

[1258] In this invention, the server includes a means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data, a means for normalizing and tagging the digital data, and a means for training a generative AI model using the normalized and tagged data. This allows maintenance response knowledge to be digitized and standardized, and efficiently shared with new technicians through the AI ​​model. It also supports real-time troubleshooting, improving the quality and efficiency of maintenance response.

[1259] A "veteran technician" is a maintenance specialist with many years of experience and advanced technical knowledge.

[1260] "Knowledge and experience in maintenance" refers to troubleshooting and problem-solving procedures and related information obtained during maintenance work.

[1261] "Digital data" refers to information such as text, images, and video that has been converted into a form that a computer can process.

[1262] "Normalization" is the process of putting data into a consistent format, which is done to ensure data integrity and consistency.

[1263] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[1264] A "generative AI model" is an artificial intelligence model that learns problem-solving procedures and efficient response methods based on collected data and automatically generates them.

[1265] A "real-time support system" is a system that allows new engineers and others to receive immediate support while performing maintenance work.

[1266] "Means for accepting inquiries" refers to an interface or device for receiving questions or problems from users.

[1267] "Means for generating response procedures" refers to methods and techniques for creating optimal solution procedures based on the content of inquiries received from users.

[1268] "Means for collecting feedback" refers to a method or system for collecting evaluations and result information from users after maintenance has been performed.

[1269] "Retraining" is the process of re-learning an AI model using new data collected to improve its accuracy and efficiency.

[1270] "Data preprocessing and formatting" refers to a series of procedures for converting collected data into a format suitable for analysis or training.

[1271] This invention is a system for speeding up and streamlining maintenance responses for storefront equipment. It utilizes generative AI models to aggregate the knowledge and experience of veterans, providing a mechanism for all staff to respond as experts. The system includes the following main components:

[1272] 1. Digitization of knowledge and experience

[1273] The server collects maintenance knowledge and experience from veteran technicians and stores it as digital data. Specifically, it includes information such as past trouble cases recorded by technicians, problem-solving procedures, and methods for identifying troubles. This digital data is stored in a database and used for later processing. The collected data is stored in text, image, and video formats.

[1274] 2. Data preprocessing and formatting

[1275] The server normalizes the collected digital data and compiles it into a unified format. For example, it converts trouble cases recorded in different formats into a unified text format. It also tags the data with relevant keywords and classification labels. This improves data searchability and enables more efficient use.

[1276] 3. Training the generative AI model

[1277] The server uses preprocessed digital data to train a generative AI model. Large amounts of historical data are used, particularly to learn troubleshooting procedures and efficient response methods. This AI model is then trained using machine learning algorithms, giving it the ability to flexibly respond to new problems.

[1278] 4. Providing real-time support

[1279] Users (engineers and new employees) receive support for maintenance responses through a dedicated application. When the user inputs the problem into the device, the device sends the input information to the server. The server analyzes the received information and generates the optimal response procedure using a generative AI model. The generated procedure is then sent to the device, which displays it to the user.

[1280] 5. Present and implement response procedures

[1281] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[1282] 6. Gather feedback and retrain

[1283] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, and response efficiency. The device then sends this feedback to the server, which processes it and reflects it in the generated AI model. The AI ​​model is retrained based on the feedback data, thereby continuously improving the accuracy and efficiency of the entire system.

[1284] Examples:

[1285] Example 1: Screen not displaying issue

[1286] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input "screen not displaying" into the terminal. The server analyzes this information and uses a generative AI model to suggest next steps:

[1287] 1. Check the power cable connection.

[1288] 2. Check that the voltage is correct.

[1289] 3. Reboot your system.

[1290] The user responds accordingly and provides feedback to the terminal that the problem has been solved.

[1291] Example 2: Communication error problem

[1292] When a user encounters a communication error, they input "communication error" into their device. The server uses a generative AI model to suggest the following steps:

[1293] 1. Check the cable connections.

[1294] 2. Restart your router.

[1295] 3. Check the server settings.

[1296] The user follows the procedure and responds by providing feedback on the final result.

[1297] As described above, this invention improves the quality and efficiency of maintenance response by consolidating the knowledge and experience of veterans and providing support in real time.

[1298] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1299] Step 1:

[1300] The server collects maintenance knowledge and experience from veteran engineers. Specifically, it collects data such as trouble cases recorded by engineers in the past, problem-solving procedures, and identification methods, and stores it in a database. The input is notes and reports on the engineers' experience and knowledge, and the output is digital data in a unified format.

[1301] Step 2:

[1302] The server normalizes and tags the collected digital data. The input is the digital data collected in step 1, and normalization and tagging improve the consistency and searchability of the data. The output is normalized and tagged data. Specifically, data in different formats is converted into a unified format and keywords and classification labels are assigned.

[1303] Step 3:

[1304] The server trains the generative AI model using the preprocessed digital data. The input is the data processed in step 2, and the machine learning algorithm is used to train the AI ​​model. The output is the trained generative AI model, which can generate troubleshooting procedures.

[1305] Step 4:

[1306] The user inputs the problem into the terminal. The specific input is a symptom of the problem, such as "the screen is not displayed" or "communication error." The terminal receives this input and sends it to the server. The input is the user's problem, and the output is the data to be sent.

[1307] Step 5:

[1308] The server analyzes the problem sent by the user and generates the optimal response procedure. The input is the problem data sent in step 4, and the response procedure is automatically generated using a generative AI model. The output is the generated response procedure. Specifically, the problem data is analyzed, and procedures such as "checking the power cable connection" and "measuring voltage" are created using the AI ​​model.

[1309] Step 6:

[1310] The server sends the generated response procedure to the terminal. The input is the response procedure generated in step 5, and the output is the data to be sent to the terminal. The specific operation is to transfer the generated procedure to the terminal.

[1311] Step 7:

[1312] The terminal displays the response procedures it has received to the user. The input is the response procedure data received in step 6, and the output is the display to the user. Specifically, the response procedures are displayed in list format on the terminal screen.

[1313] Step 8:

[1314] The user performs maintenance actions by following the procedures displayed on the terminal. The input is the procedures displayed in step 7, and the output is the results of performing the maintenance actions. The specific actions are the user performing physical operations based on the procedures.

[1315] Step 9:

[1316] The user inputs the results of the maintenance work into the terminal. Specific inputs include "the problem has been resolved" or "an additional problem has occurred." The terminal sends this feedback to the server. The input is the user's feedback, and the output is the data to send it.

[1317] Step 10:

[1318] The server processes the feedback and retrains the generative AI model. The input is the feedback data sent in step 9, and the output is the retrained generative AI model. The specific operation is to analyze the feedback data and update the AI ​​model taking into account the new data.

[1319] (Application example 1)

[1320] 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."

[1321] On-site maintenance is particularly difficult for new employees and inexperienced technicians, and requires a fast and accurate response. In addition, the robots used in factories have complex structures, so a high level of specialized knowledge is required to respond when a problem occurs. For this reason, a system is needed that allows new technicians to respond quickly and at the same level as experienced technicians.

[1322] 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.

[1323] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran engineers as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance and retraining the generative AI model; means for providing information to a dedicated device for performing maintenance on robots operating in factories; and means for displaying the maintenance response procedures to the user using the dedicated device. This enables even novice engineers to perform robot maintenance quickly and accurately.

[1324] A "veteran engineer" is an engineer who has advanced specialized knowledge and practical skills based on many years of experience.

[1325] "Maintenance response" refers to work such as inspection, repair, and adjustment carried out to maintain the normal operation of equipment and systems.

[1326] "Digital data" refers to data handled by digital devices such as computers, and includes text, images, videos, etc.

[1327] "Normalization" is the process of converting data into a unified format.

[1328] "Tagging" is the process of assigning relevant keywords and classification labels to data.

[1329] A "generative AI model" is an artificial intelligence model that is trained using machine learning algorithms based on collected data.

[1330] A "real-time support system" is a system designed to allow users to receive support in real time.

[1331] "User inquiries" refer to the act of a user reporting a question or problem regarding maintenance support to the system.

[1332] "Specialized equipment" refers to devices or equipment specifically used for this invention.

[1333] "Means provided to the user" refers to the method by which the system presents the user with the most appropriate response procedures and information.

[1334] "Feedback after maintenance" refers to evaluations and opinions provided by users after maintenance work has been completed.

[1335] "Retraining" is the process of retraining a previously trained AI model with new data to improve the model's accuracy or performance.

[1336] "Robots operating in factories" refers to automated robots used on factory production lines and in workshops.

[1337] "Maintenance response procedures" refer to the specific procedures and steps for performing maintenance work on equipment or systems.

[1338] This invention is a system that efficiently performs maintenance on robots operating in factories by aggregating the maintenance knowledge and experience collected from skilled technicians as digital data and using that data to train a generative AI model. This system includes the following main components:

[1339] 1. Digitization of knowledge and experience

[1340] The server collects maintenance knowledge and experience gathered from veteran technicians as digital data. This digital data includes past trouble cases, problem-solving procedures, and methods for identifying troubles. This creates a digital archive of knowledge.

[1341] 2. Data preprocessing and formatting

[1342] The server normalizes and tags the collected digital data. Normalization is the process of organizing data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1343] 3. Training the generative AI model

[1344] The server uses the preprocessed digital data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods from the collected data, and automatically generates optimal responses.

[1345] 4. Providing real-time support

[1346] Users (engineers and new employees) receive support during robot maintenance through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[1347] 5. Present and implement response procedures

[1348] The terminal receives the response procedures from the server and displays them to the user, allowing the user to quickly and efficiently perform maintenance. Specific procedures include detailed steps such as "check that the power cable is connected correctly" and "measure whether the voltage is appropriate."

[1349] 6. Gather feedback and retrain

[1350] After the maintenance response is completed, the user enters feedback into the device about the results, time required, and response efficiency. The device then sends this feedback to the server, which then incorporates the new feedback data into the generated AI model for retraining, thereby continuously improving the system's accuracy and efficiency.

[1351] Examples:

[1352] Example 1: Robot arm not moving

[1353] When a user (a new engineer) encounters a problem where the robot arm won't move, they input the details of the problem using a dedicated terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "check the actuator operation," and "check the system log," and provides these to the user via the terminal. The user then takes action according to these steps, and receives feedback to the terminal that the problem has been resolved.

[1354] Example prompt sentence:

[1355] "What is the best procedure to solve the problem of a stuck robot arm?"

[1356] In this way, the system of the present invention consolidates the knowledge and experience of veterans and provides real-time support, thereby improving the quality and efficiency of maintenance responses.

[1357] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1358] Step 1:

[1359] The server collects maintenance knowledge and experience from veteran engineers as digital data. Specifically, information such as past trouble cases, problem-solving procedures, and methods for identifying troubles is digitized as text data, images, and videos. The input here is the experience and knowledge of the engineers, which is converted into digital format. The output is maintenance knowledge data stored as a digital archive.

[1360] Step 2:

[1361] The server normalizes and tags the collected digital data. Normalization is the process of standardizing the data format, and tagging is the process of assigning relevant keywords and classification labels. The input is the digital data, which is normalized and tagged. The output is the unified format and tagged data.

[1362] Step 3:

[1363] The server trains the generative AI model using the preprocessed digital data. After the data preprocessing is complete, the AI ​​model is trained using a machine learning algorithm. The input is the preprocessed digital data, and the output is the trained generative AI model.

[1364] Step 4:

[1365] The server integrates the trained generative AI model into the real-time support system. The AI ​​model is incorporated into the entire system so that it can generate maintenance response procedures in real time. The input is the trained generative AI model, and the output is the generative AI model that has become part of the real-time support system.

[1366] Step 5:

[1367] The terminal accepts inquiries from users. The user inputs questions or problems related to maintenance response through the terminal (e.g., smart glasses). The input is the content of the user's inquiry. The output is the content of the inquiry being sent to the server.

[1368] Step 6:

[1369] The server analyzes the inquiry received from the user and generates the optimal response procedure based on the generative AI model. The input is the inquiry from the user, and the output is the response procedure. Specifically, the generative AI model automatically generates the procedure that is most appropriate for the trouble content.

[1370] Step 7:

[1371] The terminal displays the response procedure sent from the server to the user. The input is the response procedure sent from the server, and the output is the specific maintenance response procedure provided to the user. Specifically, the procedure is displayed on the display of the smart glasses.

[1372] Step 8:

[1373] The user actually performs the maintenance work according to the procedure. During the process, the user carries out the necessary checks and repairs based on the procedure presented. The input is the procedure provided from the terminal, and the output is the result of the maintenance work.

[1374] Step 9:

[1375] After the maintenance response is completed, the user inputs feedback on the response results, required time, and response efficiency into the terminal. The input is feedback on the results and efficiency of the maintenance response, and the output is feedback data sent from the terminal to the server.

[1376] Step 10:

[1377] The server uses the collected feedback data to retrain the generative AI model: the input is the feedback data, and the output is a new generative AI model with improved accuracy and efficiency.

[1378] As a specific example, based on the prompt, "Please tell me the best procedure to solve the problem of the robot arm not moving," the user queries the server, and the server generates a procedure for the user to check using smart glasses while carrying out maintenance work.

[1379] 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.

[1380] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system for speeding up and streamlining maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[1381] 1. Digitization of knowledge and experience

[1382] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[1383] 2. Data Preprocessing and Normalization

[1384] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1385] 3. Training the generative AI model

[1386] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[1387] 4. Providing real-time support

[1388] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[1389] 5. Use of Emotion Engine

[1390] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[1391] 6. Emotion-based adjustment of response procedures

[1392] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[1393] 7. Present and implement response procedures

[1394] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[1395] 8. Gather feedback and retrain

[1396] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[1397] Examples:

[1398] Example 1: Screen not displaying issue

[1399] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[1400] Example 2: Communication error problem

[1401] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[1402] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[1403] The processing flow will be explained below.

[1404] Step 1:

[1405] The server collects maintenance knowledge and experience from veteran technicians, including digital data on past trouble cases, resolution procedures, and methods for identifying problems. The collected data is stored in a database.

[1406] Step 2:

[1407] The server normalizes and tags the collected data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1408] Step 3:

[1409] The server uses the normalized and tagged data to train the generative AI model. During the training process, the data is fed into the generative AI model and the model's weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[1410] Step 4:

[1411] The server integrates the trained generative AI model into a real-time support system, enabling the support system to respond to user inquiries in real time.

[1412] Step 5:

[1413] When performing maintenance, users (engineers and new employees) log in to a dedicated application. The terminal displays a form to accept user input.

[1414] Step 6:

[1415] The user enters details of the problem (e.g., the screen is not displaying) into the device, which then sends this information to the server.

[1416] Step 7:

[1417] The server analyzes the received information and inputs it into the generative AI model, which then generates the optimal response procedure, which the server then sends to the device.

[1418] Step 8:

[1419] The device displays the generated troubleshooting instructions to the user. The instructions include specific steps and points to check. For example, "Check the power cable connection status" or "Measure the voltage to see if it is correct."

[1420] Step 9:

[1421] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[1422] Step 10:

[1423] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[1424] Step 11:

[1425] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[1426] Step 12:

[1427] The user performs maintenance actions by following the procedures provided, which allows the problem to be resolved quickly and efficiently.

[1428] Step 13:

[1429] After the user has resolved the problem, they input feedback into the device about the effectiveness of the procedure, the time it took, and the quality of the response, which is then sent to the server.

[1430] Step 14:

[1431] The server receives the feedback and stores it as a new dataset, which it then uses to retrain the generative AI model and emotion engine, improving the accuracy and efficiency of the system.

[1432] This series of processing steps enables maintenance of storefront equipment to be carried out quickly and efficiently, improving the overall quality of work.

[1433] Example 2

[1434] 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."

[1435] In on-site maintenance work, new and inexperienced engineers are required to solve problems quickly and efficiently on the spot. However, with conventional systems, it is difficult to share the knowledge and experience gained from veteran engineers in real time, and it is also difficult to respond flexibly while taking into account the user's emotional state. As a result, the quality and efficiency of maintenance work declines.

[1436] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1437] In this invention, the server includes: means for collecting maintenance response knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the user inquiries and generating optimal response procedures; means for providing the generated response procedures to the user; means for incorporating an emotion engine into the terminal and recognizing the emotional state of the user from facial expressions and voice; means for adjusting the response procedures based on data from the emotion engine; and means for collecting feedback after maintenance responses and retraining the generative AI model. This enables effective maintenance responses that take the user's emotional state into consideration, thereby improving the quality and efficiency of responses.

[1438] A "technician" is a person who has specialized knowledge and skills related to the maintenance, repair, or operation of a system or equipment.

[1439] "Digital data" refers to maintenance knowledge and experience expressed in an electronic format, and includes text data, images, audio data, and the like.

[1440] "Normalization" refers to the process of converting collected data into a unified format and making it consistent.

[1441] "Tagging" is the process of assigning relevant keywords and classification labels to digital data to make it easier to search and analyze later.

[1442] A "generative AI model" is an artificial intelligence model that can learn from collected and preprocessed data to perform specific tasks.

[1443] "Training" is the process of feeding data and adjusting the parameters of a generative AI model in order to learn useful knowledge and patterns.

[1444] A "real-time support system" is a system that provides immediate support for problems that arise during maintenance.

[1445] "User" refers to the engineers and technicians who use the system to perform maintenance.

[1446] The "emotion engine" is a system that analyzes the user's emotional state from their facial expressions and voice, and determines the appropriate response based on that.

[1447] "Feedback" refers to information about the results and emotional state of a user after a maintenance response.

[1448] "Retraining" is the process of retraining a generative AI model based on newly collected feedback data to improve the model's accuracy and performance.

[1449] The present invention relates to a system for quickly and effectively carrying out maintenance responses, and this system is mainly composed of three elements: a server, a terminal, and a user. How this system functions will be explained in detail below.

[1450] 1. Digitization of knowledge and experience

[1451] The server collects the maintenance knowledge and experience of veteran technicians as digital data. This includes text files, images, and audio data. For example, it collects examples of problems that technicians have solved in the past, along with detailed troubleshooting procedures. This data is stored in a database and can be used for later analysis and retrieval.

[1452] 2. Data Preprocessing and Normalization

[1453] The server converts the collected data into a unified format (normalizes it) and attaches relevant tags to it. Normalization and tagging improve the searchability and usability of the data. For example, adding tags such as "power supply trouble" or "network error" makes it easier to search later.

[1454] 3. Training the generative AI model

[1455] The server uses the preprocessed data to train the generative AI model. During this training process, a large amount of data is fed into the AI ​​model, and by adjusting the weights of the neural network, the AI ​​model learns how to resolve problems. For example, each maintenance procedure and solution method is memorized in the model.

[1456] 4. Providing real-time support

[1457] Users receive support during maintenance work using a dedicated application. The device accepts user input and sends it to a server. The server analyzes the received inquiry and generates optimal response procedures based on a generative AI model. The device then displays the generated response procedures to the user. For example, for a problem such as "the screen is not displaying," specific procedures such as "check the power cable connection" and "restart the system" are provided.

[1458] 5. Use of Emotion Engine

[1459] The device is equipped with an emotion engine that recognizes the user's emotional state from their facial expressions, voice, text, etc. For example, it uses a camera and microphone to analyze the user's facial expressions and tone of voice and extract information such as whether the user is feeling stressed or relieved.

[1460] 6. Emotion-based adjustment of response procedures

[1461] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, it adds more detailed and careful explanations. In this way, the response procedures can be dynamically adjusted according to the user's emotional state.

[1462] 7. Present and implement response procedures

[1463] The terminal displays the adjusted response procedures to the user. For example, specific steps such as "check the power cable connection" and "measure the voltage" are displayed in a list, and the user follows these to perform the maintenance work.

[1464] 8. Gather feedback and retrain

[1465] After the maintenance response is completed, the user enters feedback into the device about the results, time required, response efficiency, and emotional state. The device then sends the feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine. This continuously improves the accuracy and efficiency of the system.

[1466] Examples:

[1467] Example 1: When a user encounters a problem with a device in a store, such as "the screen is not displaying," they input the details of the problem through the device. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection," "measure the voltage," and "restart the system." The device's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and responds, and the device receives feedback that the problem has been resolved.

[1468] Example 2: When a user encounters a "communication error," they use their device to send the error details to the server. The server uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents these to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to allow the user to complete the steps more quickly.

[1469] Example prompt sentence:

[1470] Sample prompt 1: "Please describe a problem a junior engineer encounters while working on a maintenance project in-store. Also, describe how the user's emotional state is reflected when the steps to resolve the problem are displayed. For example, when an error message appears."

[1471] Such a system not only improves the quality and efficiency of maintenance responses, but also enables flexible responses that take into account the user's emotional state.

[1472] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1473] System processing flow

[1474] Step 1: Gathering knowledge and experience

[1475] Input: Knowledge and experience from veteran engineers (text files, images, audio data, etc.)

[1476] The server collects maintenance knowledge and experience from veteran technicians, specifically documents, logs, voice memos, manuals, and other data used by technicians in maintenance operations, as digital data.

[1477] How it works: The server automatically retrieves files of various formats uploaded by technicians and stores them in a database, for example, by importing data from the technician's PC or cloud storage.

[1478] Output: Knowledge and experience in digital data format is stored in a database.

[1479] Step 2: Preprocessing and normalizing data

[1480] Input: Knowledge and experience of digital data formats

[1481] The server normalizes and tags the collected digital data, converting each piece of data into a unified format and assigning relevant keywords and classification labels.

[1482] How it works: The server converts, for example, audio data into text, parses the text data using natural language processing (NLP) to put it into a consistent format, and then assigns tags like "power outage" or "network error."

[1483] Output: A normalized and tagged dataset.

[1484] Step 3: Training the generative AI model

[1485] Input: Normalized and tagged dataset

[1486] The server uses the preprocessed data to train a generative AI model, which then learns troubleshooting procedures and efficient response methods.

[1487] How it works: The server uses a neural network library (such as TensorFlow or PyTorch) to feed the tagged data to a generative AI model, adjusting the model's parameters (weights) to learn the optimal solution procedure.

[1488] Output: A trained generative AI model.

[1489] Step 4: Receiving inquiries from users

[1490] Input: User-entered problem details for maintenance

[1491] The user uses a dedicated application to input details of the problem that is currently being handled by the maintenance team. The terminal accepts this input and sends it to the server.

[1492] How it works: The user enters a specific problem into the application, such as "no screen display," and the device sends that information to the server.

[1493] Output: The query is sent to the server.

[1494] Step 5: Inquiry analysis and response procedure generation

[1495] Input: Inquiry sent by the user

[1496] The server analyzes the received inquiry and generates the optimal response procedure based on the generative AI model.

[1497] How it works: The server uses natural language processing to analyze the inquiry and generate specific response procedures such as "check the power cable connection" and "restart the system."

[1498] Output: The optimal response procedure is generated.

[1499] Step 6: View response procedures

[1500] Input: Generated response procedure

[1501] The terminal displays the generated response procedure to the user.

[1502] What it does: The device application will display instructions in text format, such as "Check the power cable. Then measure the voltage."

[1503] Output: Instructions for action are displayed to the user.

[1504] Step 7: Use the Emotion Engine

[1505] Input: User's facial expression and voice data

[1506] The terminal uses an emotion engine to recognize the user's emotional state from their facial expressions and voice.

[1507] How it works: The device analyzes images and audio captured by the camera and microphone to determine the user's emotions. For example, if the user is frowning, it will determine that the user is in a state of stress.

[1508] Output: User's emotional state data.

[1509] Step 8: Adjust response procedures

[1510] Input: User emotional state data

[1511] The server adjusts the response procedures generated by the generative AI model based on the emotional data from the emotion engine.

[1512] What it does: The server adds reassuring statements like "Don't worry" to the steps if the user is feeling anxious, or simplifies the steps if the user is feeling impatient.

[1513] Output: Coordinated response procedures.

[1514] Step 9: View the adjusted steps

[1515] Input: Coordinated response procedures

[1516] The terminal displays the adjusted procedure to the user.

[1517] What it does: The adjusted steps are displayed on the device, and the user can refer to them to perform the maintenance action. For example, the message "Don't worry, this step will be over soon."

[1518] Output: Adjusted response instructions displayed to the user.

[1519] Step 10: Gather feedback

[1520] Inputs: Maintenance response results, response time, response efficiency, emotional state

[1521] After the maintenance is completed, the user inputs feedback into the terminal.

[1522] Action: The user enters feedback such as "Problem solved", "Took 15 minutes", and "I was anxious during the process".

[1523] Output: Feedback data is sent from the device to the server.

[1524] Step 11: Retraining the generative AI model and emotion engine

[1525] Input: Collected feedback data

[1526] The server retrains the generative AI model and emotion engine based on newly collected feedback data.

[1527] How it works: The server uses the feedback data to readjust the parameters of the generative AI model to improve the accuracy and efficiency of the system, and also retrains the emotion engine to more accurately judge user emotions.

[1528] Output: Retrained generative AI model and emotion engine.

[1529] (Application example 2)

[1530] 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."

[1531] Existing maintenance support systems rely on the knowledge and experience of the technician, making it difficult for new or inexperienced technicians to provide effective maintenance support. Furthermore, support that does not take into account the user's emotional state can cause stress and anxiety. Another issue is the lack of real-time support, which means it takes a long time to resolve problems.

[1532] 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.

[1533] In this invention, the server includes: means for collecting maintenance knowledge and experience collected from veteran technicians as digital data; means for normalizing and tagging the digital data; means for training a generative AI model using the normalized and tagged data; means for integrating the trained generative AI model into a real-time support system; means for accepting inquiries from users; means for analyzing the inquiries from the users and generating optimal response procedures; means for providing the generated response procedures to the users; means for collecting feedback after maintenance responses and retraining the generative AI model; means for adjusting the generated response procedures using an emotion engine that recognizes the user's emotions; and means for collecting and analyzing the user's emotions using a camera and microphone of a smart device, thereby enabling flexible and effective real-time support according to the user's emotional state.

[1534] "Collecting knowledge and experience" is the process of collecting knowledge and experience regarding maintenance responses from veteran technicians as digital data.

[1535] "Normalization and tagging" is the process of converting collected digital data into a unified format and assigning relevant keywords and classification labels.

[1536] "Training a generative AI model" is the process of using normalized and tagged data to teach a generative AI model how to respond efficiently.

[1537] "Real-time support system integration" is the process of building a system that uses trained generative AI models to provide support to users in real time.

[1538] "Accepting inquiries from users" is a means by which users can send questions about maintenance responses or trouble reports to the system.

[1539] "Analysis of inquiries and generation of response procedures" is a process of analyzing an inquiry from a user and generating an optimal response procedure based on that analysis.

[1540] An "emotion engine" is an engine that recognizes and analyzes a user's emotional state from facial expressions, voice, text, etc.

[1541] "Use of smart devices" refers to the use of devices such as smartphones and tablets to collect the user's emotional state through cameras and microphones.

[1542] "Feedback collection and retraining" refers to the process of collecting feedback from users after maintenance is performed and retraining the generative AI model based on that feedback.

[1543] This invention combines a generative AI model with an emotion engine that recognizes user emotions in a system that speeds up and streamlines maintenance of storefront equipment. This system is composed of the following main components and processing flow:

[1544] 1. Digitization of knowledge and experience

[1545] The server collects the maintenance knowledge and experience gathered from veteran technicians as digital data, including past trouble cases, problem-solving procedures, and methods for identifying troubles. The collected data is stored in a database.

[1546] 2. Data Preprocessing and Normalization

[1547] The server normalizes and tags the collected digital data. Normalization is the process of converting data into a unified format, and tagging is the process of assigning relevant keywords and classification labels to the data, making it more searchable and useful.

[1548] 3. Training the generative AI model

[1549] The server uses the preprocessed data to train the generative AI model. During the training process, the data is fed to the generative AI model and the model weights are adjusted, allowing the model to learn troubleshooting procedures and efficient response methods.

[1550] 4. Providing real-time support

[1551] Users (engineers and new employees) receive support during maintenance operations through a dedicated application. The terminal accepts user input and sends it to the server. The server analyzes the received inquiry, generates optimal response procedures based on a generative AI model, and sends them to the terminal. The terminal then displays the generated response procedures to the user.

[1552] 5. Use of Emotion Engine

[1553] The device is equipped with an emotion engine that recognizes the user's emotional state from facial expressions, voice, text, etc. For example, the emotional state can be detected by analyzing the user's facial expressions and tone of voice using a camera or microphone. The emotion engine analyzes this data in real time and extracts information such as whether the user is feeling stressed or relieved.

[1554] 6. Emotion-based adjustment of response procedures

[1555] The server adjusts the response procedures generated by the generative AI model based on the user's emotional data sent from the emotion engine. For example, if the user is nervous, the server will adjust the procedures by adding more detailed and careful explanations. This allows the server to provide more effective support according to the user's emotional state.

[1556] 7. Present and implement response procedures

[1557] The terminal displays the adjusted response procedures to the user. Specific steps and points to check are displayed in a list, and the user follows them to perform maintenance. For example, steps such as "check the power cable connection status" and "measure whether the voltage is appropriate" are presented.

[1558] 8. Gather feedback and retrain

[1559] After the maintenance response is completed, the user enters feedback into the device regarding the results, time required, response efficiency, and emotional state. The device then sends this feedback to the server, which then incorporates the new feedback data into the generative AI model and retrains it, including the emotional engine, thereby continuously improving the accuracy and efficiency of the system.

[1560] Examples:

[1561] Example 1: Screen not displaying issue

[1562] When a user (a new engineer) encounters a problem with the device's screen not displaying in a store, they input the details of the problem using a terminal. The server analyzes the information and uses a generative AI model to generate steps such as "check the power cable connection status," "measure the voltage," and "restart the system," which are provided to the user via the terminal. The terminal's emotion engine detects that the user looks anxious, and the server adds a reassuring statement to the steps, such as "Don't worry, the next step is easy." The user follows this and takes action, and the terminal receives feedback that the problem has been resolved.

[1563] Example 2: Communication error problem

[1564] When a user encounters a communication error, they use their device to send the error details to the server. The server then uses a generative AI model to generate instructions such as "check the cable connection," "restart the router," and "check the server settings," and presents them to the user via their device. The emotion engine detects that the user is frustrated, and the server simplifies the instructions to help the user complete the steps more quickly.

[1565] In this way, the system of the present invention can further improve the quality and efficiency of maintenance responses by providing flexible and effective support that also takes into account the user's emotional state.

[1566] Example prompt sentence:

[1567] If the user types "screen not showing":

[1568] User is feeling anxious. Provide detailed and comforting support message. Base message: "Please check the connection of the power cable of the screen."

[1569] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1570] Step 1:

[1571] The terminal accepts the contents of a maintenance inquiry from the user as input. The user inputs the inquiry into the terminal, and the terminal transmits the inquiry contents to the server as text data.

[1572] Step 2:

[1573] The server analyzes the inquiry received from the device. The server uses a generative AI model to analyze the inquiry and generate appropriate response procedures. The server uses the text data of the inquiry as input and provides the generated text of the response procedures as output.

[1574] Step 3:

[1575] The terminal provides the user with the response procedure sent from the server, and receives the generated response procedure text from the server and displays it to the user in an appropriate format.

[1576] Step 4:

[1577] The device uses an emotion engine to collect the user's emotional state as input. It uses a camera and microphone to analyze the user's facial expressions and tone of voice in real time to obtain emotion data. It uses camera images and audio data as input and obtains emotional state data as output.

[1578] Step 5:

[1579] The server adjusts the response procedure created by the generative AI model based on the acquired emotional state data. It receives the emotional state data as input and inputs the prompt sentence into the generative AI model to create a newly adjusted response procedure. As an output, it obtains the text of the response procedure adjusted based on the emotion.

[1580] Step 6:

[1581] The terminal displays the adjusted response procedure to the user. The terminal receives the text of the adjusted response procedure from the server and displays it to the user.

[1582] Step 7:

[1583] The user performs maintenance and inputs the results as feedback to the terminal. After the maintenance is completed, the results, required time, emotional state, etc. are input to the terminal, and the terminal sends the contents to the server. The text data of the feedback content is used as input, and feedback data is provided to be sent to the server as output.

[1584] Step 8:

[1585] The server retrains the generative AI model using the received feedback data. It receives the feedback data as input, adjusts the weights of the generative AI model based on the feedback data, and retrains the model. It provides an updated generative AI model as output.

[1586] 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.

[1587] 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.

[1588] 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.

[1589] 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.

[1590] FIG. 9 is a diagram illustrating 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 actions 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.

[1591] 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.

[1592] 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).

[1593] 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.

[1594] 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."

[1595] 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.

[1596] 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).

[1597] 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.

[1598] 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.

[1599] 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.

[1600] 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.

[1601] 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.

[1602] 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.

[1603] 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.

[1604] 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.

[1605] 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.

[1606] 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.

[1607] The following is further disclosed regarding the above embodiment.

[1608] (Claim 1)

[1609] A means of collecting maintenance knowledge and experience gathered from veteran engineers as digital data,

[1610] means for normalizing and tagging said digital data;

[1611] means for training a generative AI model using the normalized and tagged data;

[1612] A means for integrating the trained generative AI model into a real-time support system; and

[1613] means for accepting inquiries from users;

[1614] means for analyzing the inquiry from the user and generating an optimal response procedure;

[1615] means for providing the generated response procedure to a user;

[1616] A means for collecting feedback after maintenance and retraining the generative AI model.

[1617] A system including:

[1618] (Claim 2)

[1619] 2. The system of claim 1, wherein the digital data includes text data, images, and videos.

[1620] (Claim 3)

[1621] 2. The system according to claim 1, further comprising means for accepting an inquiry from the us...

Claims

1. A means of collecting maintenance knowledge and experience gathered from veteran engineers as digital data, means for normalizing and tagging said digital data; means for training a generative AI model using the normalized and tagged data; A means for integrating the trained generative AI model into a real-time support system; and means for accepting inquiries from users; means for analyzing the inquiry from the user and generating an optimal response procedure; means for providing the generated response procedure to a user; A means for collecting feedback after maintenance and retraining the generative AI model. A system including:

2. 2. The system according to claim 1, wherein the digital data includes text data, images, and videos.

3. 2. The system according to claim 1, further comprising means for accepting an inquiry from the user using a dedicated application.

Images