system

Abstract

Provide a system. [Solution] A means of storing and making searchable past approval and guideline data; a display means for displaying the past approval request data and guideline data so that the user can refer to them on the terminal where the approval request is submitted; A means to automatically check the contents of the user's approval request and suggest corrections and points to check, A means for rechecking the content of the approval request corrected by the user based on the corrections and confirmation points; A system that includes a means to send the final approval content to a higher approver.

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] The conventional process of drafting approval requests involves a lot of manual work, and referring to past approval request data and guidelines takes time, which can lead to problems with reduced efficiency. In addition, there is a tendency for corrections and confirmation work to increase due to incomplete or incorrect content in the approval request, which can cause delays in the final approval process. The objective of this invention is to solve these problems and improve the efficiency and quality of approval request drafting work. [Means for solving the problem]

[0005] The present invention solves the above problem with a system that includes a means for storing and making searchable past approval request data and guideline data, a means for displaying the past approval request data and guideline data so that it can be referenced on the terminal where the user drafts the approval request, a means for automatically checking the content of the user's approval request and presenting corrections and points to be confirmed, a means for rechecking the approval request content revised by the user based on the corrections and confirmation points, and a means for sending the final approval request content to a superior approver. Furthermore, by adopting a means for organizing and indexing past approval request data and guideline data by the same classification and a means for presenting detailed correction suggestions and points to be noted based on the results of the automatic check when the user drafts the approval request, an efficient and accurate approval request creation process is realized.

[0006] "Past approval data" refers to records of approvals that have been previously created and approved, including relevant information (project name, purpose, budget, applicant, approval date, classification, etc.).

[0007] "Guideline data" refers to data that details the standards, guidelines, approval criteria, points to note, etc. when creating a request for approval.

[0008] "Means of storing and making it searchable" refers to technology that allows past approval requests and guideline data to be stored in storage such as a database, and for this data to be searched and retrieved efficiently.

[0009] "User" refers to the individual or employee who creates and submits a request.

[0010] "Terminal" refers to a device such as a computer or mobile terminal that a user uses to draft an approval request.

[0011] "Means for displaying" refers to technology that allows past approval requests and guideline data to be visually viewed on the user's device.

[0012] "Means for automatically checking and suggesting corrections and points to check" refers to technology that analyzes the content of the approval request entered by the user, checks the accuracy and completeness of the content based on specific standards and rules, and presents the results to the user.

[0013] "Means for rechecking" refers to technology that allows the user to recheck the contents of the approval request after making corrections and present the results of the recheck.

[0014] "Means for sending the final approval request to a higher-level approver" refers to the technology for sending the approval request after all checks and corrections have been completed to the higher-level person or administrator who is responsible for approval.

[0015] "Means of organizing and indexing by classification" refers to technology that organizes past approval data and guideline data based on categories and classifications, and indexes them so that they can be searched efficiently.

[0016] "Means for providing detailed suggestions for corrections and points to be aware of" refers to technology for providing specific suggestions for corrections and points to be aware of as feedback to the user based on the results of the check. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

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

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

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

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0032] 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.

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

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] The embodiments of the present invention will be specifically described below.

[0039] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process. Below, we will explain the procedure for executing the program of this system and its operation with concrete examples.

[0040] System configuration

[0041] 1. Server:

[0042] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[0043] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[0044] Check results and correction suggestions are generated and sent to the user's device.

[0045] 2. Terminal:

[0046] The terminal provides a form for users to draft a request for approval.

[0047] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[0048] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[0049] 3. User:

[0050] Users use their devices to draft requests and revise them based on feedback from the server.

[0051] The final revised approval request is sent to the higher approver.

[0052] Program processing flow

[0053] 1. Prepare the database

[0054] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[0055] 2. Start of proposal process

[0056] The user opens the proposal form on their device and enters details such as the proposal title, purpose, content, budget, and schedule.

[0057] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[0058] 3. Display of past approval requests and guidelines

[0059] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends the data to the terminal.

[0060] The terminal displays this data for the user to refer to.

[0061] 4. Run an automatic check

[0062] After the user has finished writing the details of the request, they click the "Check" button on the terminal.

[0063] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[0064] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[0065] 5. Review and correct feedback

[0066] The user checks the feedback from the server on their device and makes corrections to the request based on the points pointed out and suggestions.

[0067] After making the corrections, click the "Check" button again to request the server to check again.

[0068] 6. Final application execution

[0069] The server will recheck to make sure all check items are cleared.

[0070] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[0071] The server sends the final approval to the higher approver, and the approval process begins.

[0072] Specific examples

[0073] As an example, consider the case where a user submits a request for approval for a "new product development project."

[0074] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[0075] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0076] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[0077] 4. After filling out the form, Tanaka clicks the "Check" button.

[0078] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[0079] 6. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[0080] 7. Click the "Check" button again to request a recheck of the server.

[0081] 8. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[0082] 9. The server sends the request to the higher approver and the approval process is completed.

[0083] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests.

[0084] The processing flow will be explained below.

[0085] Step 1:

[0086] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[0087] Step 2:

[0088] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[0089] Step 3:

[0090] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[0091] Step 4:

[0092] The server receives the request and retrieves data related to the target classification from a database and returns it.

[0093] Step 5:

[0094] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[0095] Step 6:

[0096] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[0097] Step 7:

[0098] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[0099] Step 8:

[0100] The server generates the analysis results and sends feedback to the device, including check points and suggested modifications.

[0101] Step 9:

[0102] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[0103] Step 10:

[0104] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[0105] Step 11:

[0106] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[0107] Step 12:

[0108] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[0109] Step 13:

[0110] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[0111] Example 1

[0112] 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."

[0113] Conventional approval drafting systems were unable to effectively utilize past approval data or guideline data, and the quality of automatic checks and feedback on approval content was limited, resulting in issues such as insufficient efficiency and quality improvement in approval drafting.In addition, there was insufficient support for users when revising approval content, and the final confirmation and higher-level approval process was often done manually, making the entire process time-consuming and labor-intensive.

[0114] 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.

[0115] In this invention, the server includes means for storing and searching past approval request data and guideline data, means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to a superior approver, means for analyzing the approval request content using an AI model and checking the accuracy and completeness of the content based on specific standards and rules, and means for providing the user with specific feedback based on the analysis results of the AI ​​model, thereby enabling the efficiency and quality of approval request drafting to be improved.

[0116] "Past approval data" refers to information on approval documents created in the past, including the approval classification, project name, purpose, budget, applicant, approval date, etc.

[0117] "Guideline data" is data that includes guidelines and regulations regarding the creation of approval documents established by companies and organizations, and describes specific procedures and standards.

[0118] "Terminal" refers to an electronic device that a user operates to draft a request for approval, including a PC, tablet, smartphone, etc.

[0119] A "server" refers to a computer system that manages a database and searches and processes data in response to user requests.

[0120] "Means of making something searchable" refers to the functionality for efficiently searching for information in a database and quickly obtaining the required data.

[0121] "Means for displaying" refers to the ability to visually present data on a terminal for a user to view.

[0122] "Means for automatic checking" refers to a function in which the system analyzes the content of the approval request entered by the user and verifies the accuracy and completeness of the content.

[0123] "Means for suggesting corrections and points to check" refers to a function that points out deficiencies and errors to the user based on the results of the automatic check and notifies them of points that need to be corrected.

[0124] "Means to check again" refers to a function that allows the user to re-analyze the contents of the revised approval request and reconfirm whether the confirmation items have been cleared.

[0125] "Superior approver" refers to an individual or departmental official who has the authority to approve the final contents of the request.

[0126] "Means of analyzing the contents of approval requests using AI models" refers to a function that uses artificial intelligence technologies such as machine learning and natural language processing to perform detailed analysis of approval requests.

[0127] "Means of providing specific feedback" refers to a function that presents specific suggestions for correction and points of attention to the user based on the results of the AI ​​model's analysis.

[0128] The present invention relates to a system for improving the efficiency and quality of approval drafting through collaboration between a server, a terminal, and a user. A method for specifically implementing the program of this system will be described below.

[0129] System Overview

[0130] This system involves collaboration between the server, terminals, and users to streamline and improve the quality of approval drafting. The server uses a database management system (DBMS) to manage past approval drafting data and guideline data, and analyzes the content of approval drafts sent by users. The analysis results are fed back to the user's terminal. The terminal provides an interface for users to draft approval drafts and displays data retrieved from the server. Users use their terminals to draft approval drafts and revise the content based on feedback from the server.

[0131] Hardware and Software Configuration

[0132] Server: The server manages the database using a DBMS such as MySQL (registered trademark) or PostgreSQL. It also uses a machine learning model (AI model) for analysis.

[0133] Device: Devices include PCs, tablets, smartphones, etc. The user interface is provided via a web browser or dedicated application.

[0134] User: The user operates the terminal to draft a request for approval and revise the content based on feedback from the server.

[0135] A concrete explanation of how the system works

[0136] 1. Prepare the database:

[0137] Server: The server stores past approval data and guideline data in a DBMS, including approval classification, project name, purpose, budget, applicant, approval date, etc. The data is indexed to enable efficient searches.

[0138] 2. Initiating a proposal:

[0139] User: The user opens the request draft form on their device and enters details such as the request title, purpose, content, budget, and schedule.

[0140] Terminal: The terminal sends a request to the server based on the "category" field entered by the user.

[0141] 3. View past approvals and guidelines:

[0142] Server: Receives the request, searches for past approval data and guideline data related to the target category, and sends it to the terminal. For example, execute the following query: "SELECT FROM Ringi WHERE category = 'New product development project'"

[0143] Terminal: The terminal displays this data for the user to refer to.

[0144] 4. Run the automatic check:

[0145] User: After filling out the details of the request, click the "Check" button on the device.

[0146] Server: The server receives the request and automatically analyzes it using AI models to check the accuracy and completeness of the content based on specific criteria and rules.

[0147] Example: "Please check if this request complies with the guidelines. Content: {Request content}"

[0148] Server: Generates feedback including points that need to be corrected and points to be noted, and sends this to the device.

[0149] 5. Review and correct feedback:

[0150] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[0151] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[0152] 6. Final application execution:

[0153] Server: Recheck and make sure all checks are cleared.

[0154] User: After receiving feedback from the server that "all items are acceptable," clicks the final submission button.

[0155] Server: The final approval request is sent to the higher approver, and the approval process begins.

[0156] Specific examples

[0157] As an example, consider the case of drafting a request for approval for a "new product development project."

[0158] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[0159] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0160] 3. The user uses this data to fill out the details of the request (purpose, required budget, schedule, etc.).

[0161] 4. After filling out the form, the user clicks the "Check" button.

[0162] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[0163] 6. The user checks the feedback and makes corrections, such as specifying budget items.

[0164] 7. Click the "Check" button again to request a recheck of the server.

[0165] 8. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[0166] 9. The server sends the request to the higher approver, and the approval process begins.

[0167] As described above, by implementing the system of the present invention, the efficiency and quality of approval drafting work can be improved.

[0168] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0169] Step 1:

[0170] Preparing the database

[0171] Server: A database management system (DBMS) is used to store past approval data and guideline data in a database. Specifically, DBMSs such as MySQL and PostgreSQL are used.

[0172] Input: Existing approval data and guideline data.

[0173] Data processing: Data indexing, classification, and import.

[0174] Output: A searchable database.

[0175] Specific operation: Executes SQL queries to store approval data and guideline data, and sets up indexes.

[0176] sql

[0177] CREATE INDEX idx_category ON Ringi(category);

[0178] Step 2:

[0179] Start of drafting a request

[0180] User: Uses a device to open the request draft form and enter details such as the request title, purpose, content, budget, and schedule.

[0181] Input: Title, purpose, content, budget, schedule, etc. of the request.

[0182] Output: A completed decision proposal form.

[0183] Specific actions: Enter data into an input form in a web browser or dedicated application. Input fields include, for example, "title," "purpose," and "content."

[0184] Step 3:

[0185] Display of past approval requests and guidelines

[0186] Terminal: Sends requests to the server based on the user's input data (especially the classification fields).

[0187] Input: The value of the classification field.

[0188] Output: Request information.

[0189] Specific operation: Encodes form data in JSON format and sends it to the server.

[0190] json

[0191] { "category": "New product development project"}

[0192] Server: Receives the request and searches the database for past approval requests and guideline data related to the target classification.

[0193] Input: Request information.

[0194] Data operations: Executing SQL queries.

[0195] Output: Past approval data and guideline data.

[0196] Specific behavior: Executes an SQL query and retrieves relevant data.

[0197] sql

[0198] SELECT FROM Ringi WHERE category = 'New product development project';

[0199] Terminal: Displays the retrieved data to the user.

[0200] Input: Data from the server.

[0201] Output: Past approvals and guidelines data displayed to the user.

[0202] Specific operation: Analyze the acquired data and display it in the user interface.

[0203] Step 4:

[0204] Running automatic checks

[0205] User: After filling out all the details of the request, click the "Check" button on the device.

[0206] Input: Completed request.

[0207] Output: The request to the server.

[0208] Specific behavior: Sends form data in JSON format to the server.

[0209] Server: Analyzes the received approval request using an AI model, checking the accuracy and completeness of the content based on specific criteria and rules.

[0210] Input: Data of the request.

[0211] Data calculation: Analysis using AI models.

[0212] Output: Feedback information.

[0213] Specific operations: Generate prompt sentences for the AI ​​model and perform analysis.

[0214] Example prompt: "Please check if this request complies with the guidelines. Content: {Request content}"

[0215] Server: Generates feedback including points that need to be corrected and points to be noted, and sends it to the device.

[0216] Input: Analysis results from the AI ​​model.

[0217] Data processing: Feedback generation.

[0218] Output: Feedback to the user.

[0219] Specific operation: Sends feedback data in JSON format to the device.

[0220] Step 5:

[0221] Review and correct feedback

[0222] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[0223] Input: Feedback from the server.

[0224] Output: Revised version of the approval request.

[0225] Specific actions: Rewrite and revise based on the feedback.

[0226] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[0227] Input: Revised version of the approval request.

[0228] Output: Re-request to the server.

[0229] Specific action: Send the corrected data to the server again.

[0230] Step 6:

[0231] Final application execution

[0232] Server: Recheck and make sure all checks are cleared.

[0233] Input: Revised version of the approval request.

[0234] Data calculation: Recheck process.

[0235] Output: Final feedback.

[0236] Specific actions: Use the AI ​​model again to confirm the content.

[0237] User: After receiving feedback from the server that "All items are acceptable," clicks the final submission button.

[0238] Input: Final feedback.

[0239] Output: Final application request.

[0240] Specific operation: Click the final application button to send it to the server.

[0241] Server: The final approval request is sent to the higher approver, and the approval process begins.

[0242] Input: Final approval content.

[0243] Data processing: Generate notifications to higher approvers.

[0244] Output: Notification to approver.

[0245] Specific behavior: Sends email or system notification to the senior approver and starts the approval process.

[0246] Through the above steps, the system of the present invention realizes efficient and quality improvement in drafting approval requests.

[0247] (Application example 1)

[0248] 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."

[0249] Currently, factories require a lot of time and effort to draft and review approval requests, resulting in problems such as reduced work efficiency and inconsistent quality. Furthermore, the review process, which involves referencing past approval data and guidelines, is often done manually, increasing the risk of human error. Furthermore, there is a need to automate these processes and significantly improve efficiency by incorporating the latest technology.

[0250] 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.

[0251] In this invention, the server includes a means for storing and searching past approval request data and guideline data, a means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, a means for automatically checking the content of the user's approval request and presenting corrections and confirmation points, a means for rechecking the approval request content revised by the user based on the corrections and confirmation points, a means for sending the final approval request content to a superior approver, a means for automatically drafting and confirming approval requests during the factory production process, a means for analyzing the content of the user's approval request using a generative AI model and generating correction proposals based on prompts, a terminal display means for confirming the generated correction proposals, and a means for automatically sending the approval request content to the approver after final confirmation, thereby enabling the efficiency and quality of approval request drafting and confirmation work to be improved.

[0252] "Past approval data" is a database that stores information about approvals that have been previously submitted and approved.

[0253] "Guideline data" is a database that stores information that indicates guidelines and standards for drafting approval requests.

[0254] "Means of making something searchable" refers to the ability to quickly find the desired information from a database.

[0255] "Proposal terminal" means the electronic device used by a user to create and edit a proposal.

[0256] "Means to automatically check and suggest corrections and points to check" refers to a function that analyzes the contents of the request based on an algorithm and notifies the user of problems and areas that need to be corrected.

[0257] "Means to check again" refers to a function that reanalyzes the contents of the approval request after the user has made corrections and checks whether the problem has been resolved.

[0258] "Means to send to a higher approver" refers to the function for transferring the contents of the request that ultimately require approval to the decision maker.

[0259] "Means for automatically drafting and confirming approval requests during the factory production process" refers to a function that automatically generates approval request content and confirms that content on the factory production line.

[0260] A "generative AI model" is an algorithmic model that uses machine learning to generate and analyze text.

[0261] The "means for generating correction suggestions based on prompt sentences" refers to a function that automatically presents optimal correction suggestions based on user input.

[0262] "Terminal display means" refers to a screen or monitor for displaying information referenced by the user.

[0263] As an embodiment of the present invention, the configuration of the following system and the processing of its programs will be specifically described. Below, we will explain a method for achieving efficiency and quality improvement in drafting and checking approval requests in the production process of a factory through cooperation between a server, a terminal, and a user.

[0264] System configuration

[0265] 1. Server:

[0266] The server stores past approval data and guideline data and manages the database so that they can be searched and referenced efficiently.

[0267] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[0268] A generative AI model is used to analyze the content of the user's request and generate suggested revisions based on the prompt text.

[0269] Check results and correction suggestions are generated and sent to the user's device.

[0270] 2. Terminal:

[0271] The terminal provides a form for users to draft a request for approval.

[0272] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[0273] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[0274] It has a display means for displaying the generated correction suggestions and points to note.

[0275] 3. User:

[0276] Users use their devices to draft requests and revise them based on feedback from the server.

[0277] The final revised approval request is sent to the higher approver.

[0278] Hardware and software used

[0279] Hardware

[0280] Factory robots

[0281] Smart devices (e.g., tablets)

[0282] server

[0283] software

[0284] Database Management Systems

[0285] Natural Language Processing Library

[0286] Generative AI models (e.g., ChatGPT®)

[0287] Data processing and calculation

[0288] Database preparation

[0289] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[0290] Start of drafting a request

[0291] The user opens the approval proposal form on the terminal and enters details such as the title, purpose, content, budget, schedule, etc. The terminal then sends a request to the server based on the "category" field entered by the user when submitting the proposal.

[0292] Display of past approval requests and guidelines

[0293] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends it to the terminal. The terminal displays this data for the user to refer to.

[0294] Running automatic checks

[0295] After the user has finished filling out the request details, they click the "Check" button on their device. The server receives the request details and automatically analyzes them. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. A generative AI model is used to generate correction suggestions based on the prompt. As an example, the prompt used is, "I would like to create a request for approval regarding the introduction of a new production line. Please enter the purpose, required budget, and schedule."

[0296] Review and correct feedback

[0297] The user checks the feedback from the server on their device and corrects the approval request based on the points pointed out and suggestions. After correcting the corrections, they click the "Check" button again to request a recheck from the server.

[0298] Final application execution

[0299] The server rechecks and confirms that all check items have been cleared. The user receives feedback from the server that "all items are appropriable" and clicks the final application button. The server then sends the final approval request to the higher approver, and the approval process begins.

[0300] This will enable the efficiency of drafting and confirming approvals during the factory production process and improve quality.

[0301] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0302] Step 1:

[0303] The server stores and indexes past approval and guideline data in a database, making it possible to efficiently search the data. The server periodically updates this data to keep it up to date. The input is existing approval and guideline data, and the output is an indexed database.

[0304] Step 2:

[0305] The user opens the approval proposal form on the terminal and enters details such as the approval title, purpose, content, budget, schedule, etc. The terminal receives the user's input and sends the input data to the server. The input is the user's approval details, and the output is the data sent to the server.

[0306] Step 3:

[0307] The server receives the approval data entered by the user and searches the database for related past approval data and guideline data. The server sends the search results to the terminal and instructs it to display them. The input is the user's approval data and the database data, and the output is the search results for related past approval data and guideline data.

[0308] Step 4:

[0309] The terminal displays past approval data and guideline data received from the server for the user's reference. The user inputs and modifies the details of the approval while referring to the displayed data. The input is data from the server, and the output is displayed to the user.

[0310] Step 5:

[0311] After the user has finished entering the details of the approval request, they click the "Check" button on their device. The device then sends the user's approval request to the server. The input is the user's check request and the approval request details, and the output is the data sent to the server.

[0312] Step 6:

[0313] The server analyzes the received approval request and automatically checks it using a generative AI model. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The server generates correction suggestions based on the prompt text and sends the results to the terminal. The input is the user's approval request content and the generative AI model, and the output is correction suggestions and check results.

[0314] Step 7:

[0315] The terminal displays the correction suggestions and check results received from the server for the user to review. The user then corrects the approval request based on the feedback provided. The input is data from the server, and the output is the display and feedback to the user.

[0316] Step 8:

[0317] After the user has finished making the corrections, they click the "Check" button again, and the terminal sends the approval request content to the server again. The input is the approval request content after the user's corrections, and the output is the re-transmission to the server.

[0318] Step 9:

[0319] The server analyzes the approval request content again and checks whether all check items have been cleared. If the analysis results are satisfactory, it generates feedback stating "All items can be approved" and sends it to the terminal. The input is the revised approval request content and the generation AI model, and the output is the final feedback.

[0320] Step 10:

[0321] The user checks the final feedback and, if there are no problems, clicks the "Final Application" button. The terminal then sends the final approval request to the server. The input is the user's approval request and the final approval request, and the output is the final data sent to the server.

[0322] Step 11:

[0323] The server sends the received final approval content to the superior approver and starts the approval process. The input is the final approval content, and the output is sending the data to the superior approver and starting the approval process.

[0324] In this way, the process of drafting and reviewing approval requests is made more efficient, and quality is improved.

[0325] 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.

[0326] The embodiments of the present invention will be specifically described below.

[0327] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process.The present invention also includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[0328] System configuration

[0329] 1. Server:

[0330] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[0331] The server analyzes the content of the approval request sent by the user and automatically checks it based on guidelines and past approval request data.

[0332] Check results and correction suggestions are generated and sent to the user's device.

[0333] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[0334] 2. Terminal:

[0335] The terminal provides an input form for the user to draft a request for approval.

[0336] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[0337] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[0338] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[0339] 3. User:

[0340] Users use their devices to draft requests and revise them based on feedback from the server.

[0341] The final revised approval request is sent to the higher approver.

[0342] The emotion engine receives input through the device's camera and sensors.

[0343] Program processing flow

[0344] 1. Prepare the database

[0345] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes it for efficient searches.

[0346] 2. Start of proposal process

[0347] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[0348] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[0349] 3. Display of past approval requests and guidelines

[0350] The server receives the request and retrieves data related to the subject's classification from a database and returns it.

[0351] The terminal displays this data for the user to refer to.

[0352] 4. Run an automatic check

[0353] After the user has finished writing the details of the request, he / she clicks the "Check" button.

[0354] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[0355] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[0356] 5. Feedback adjustment

[0357] The device's emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. For example, if the user is feeling stressed, the device will provide more polite and detailed feedback.

[0358] The server adjusts the content and tone of the feedback based on information from the emotion engine.

[0359] 6. Reviewing and Correcting Feedback

[0360] The user checks the feedback from the server on the device, confirms the points pointed out and points to note, and makes any necessary corrections.

[0361] After making the corrections, click the "Check" button again to request the server to check again.

[0362] 7. Final application execution

[0363] The server will recheck to make sure all checkpoints are cleared.

[0364] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[0365] The server sends the final approval to the higher approver, and the approval process begins.

[0366] Specific examples

[0367] As an example, consider the case where a user submits a request for approval for a "new product development project."

[0368] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[0369] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0370] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[0371] 4. After filling out the form, Tanaka clicks the "Check" button.

[0372] 5. The server analyzes the request and returns feedback saying, "There is insufficient budget item detail."

[0373] 6. The device's emotion engine detects that Tanaka is feeling a little stressed.

[0374] 7. The server then generates more detailed suggestions for correction and feedback in a gentler tone and sends it to Tanaka.

[0375] 8. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[0376] 9. Click the "Check" button again to request a recheck of the server.

[0377] 10. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[0378] 11. The server sends the request to the higher approver and the approval process is completed.

[0379] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests, taking into account the emotional state of the user.

[0380] The processing flow will be explained below.

[0381] Step 1:

[0382] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[0383] Step 2:

[0384] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[0385] Step 3:

[0386] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[0387] Step 4:

[0388] The server receives the request and retrieves data related to the target classification from a database and returns it.

[0389] Step 5:

[0390] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[0391] Step 6:

[0392] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[0393] Step 7:

[0394] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[0395] Step 8:

[0396] To recognize the user's emotional state, the device uses cameras and sensors to analyze the user's facial expressions and tone of voice using an emotion engine.

[0397] Step 9:

[0398] The server generates the analysis results and adjusts the feedback based on the information from the emotion engine, for example, if the user is feeling stressed, the feedback will be presented in a gentler tone.

[0399] Step 10:

[0400] The server generates the check results and correction suggestions and sends them to the terminal.

[0401] Step 11:

[0402] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[0403] Step 12:

[0404] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[0405] Step 13:

[0406] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[0407] Step 14:

[0408] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[0409] Step 15:

[0410] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[0411] Example 2

[0412] 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."

[0413] In conventional approval proposal systems, users had to take a lot of time and effort to submit approvals, and the quality of the approvals was inconsistent. This could also cause stress and strain on users, making it difficult to create efficient, high-quality approval proposals. Furthermore, because feedback was uniform, flexible support that took into account the user's emotional state was not possible.

[0414] 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.

[0415] In this invention, the server includes a means for storing and retrieving past approval data and guideline data, a means for automatically checking the content of a user's approval and suggesting corrections and points to check, and a means for recognizing the user's emotional state and adjusting the method and content of feedback. This improves the efficiency of the user's approval drafting process and the quality of approval content. Furthermore, it is possible to reduce the user's stress and burden and provide flexible support according to the user's emotional state.

[0416] "Past approval data" refers to information related to approval documents that have previously been created or approved, which is stored in a database and is searchable.

[0417] "Guideline data" refers to documents and data that describe standard or recommended procedures, conditions, and standards for drafting approvals.

[0418] "Classification" refers to categories for organizing approval data and guideline data by purpose, project, or other criteria.

[0419] "Automatic checking" refers to the process in which the system analyzes the contents of a user's request and verifies it against pre-defined standards and rules.

[0420] "Points to correct and check" indicate the parts of the user's approval request that need to be corrected or items that should be checked, as identified as a result of the automated check.

[0421] "Emotional state" refers to the user's current psychological or emotional state, and is estimated from facial expressions, tone of voice, and the like.

[0422] "Feedback" refers to information and advice provided to the user by the system, which is useful for correcting and improving the content of the request.

[0423] "Superior approver" refers to the person who has the authority to make the final evaluation of the contents of the request and decide whether to approve or reject it.

[0424] "Indexing" refers to the process of assigning search keys to specific items so that information in a database can be efficiently searched.

[0425] A specific embodiment for carrying out the present invention will now be described. The system of the present invention works in cooperation with a server, a terminal, and a user to improve the efficiency and quality of drafting approval requests. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it also includes a function that provides flexible support.

[0426] System configuration

[0427] 1. Server:

[0428] The server manages a searchable database that stores past approval and guideline data, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data to enable efficient searches.

[0429] The server analyzes the content of the approval request submitted by the user and automatically checks it based on guidelines and past approval request data. Specifically, it uses a generative AI model to check the accuracy and completeness of the content.

[0430] The server generates the check results and correction suggestions and sends them to the user's device. This feedback is provided in a way that takes into account the user's emotional state. The content and tone of the feedback are adjusted based on information from the emotion engine.

[0431] Send the final approval request to a higher approver to begin the approval process.

[0432] 2. Terminal:

[0433] The terminal provides an input form for the user to submit a request for approval, which includes fields for entering details such as the title, purpose, content, budget, and schedule of the request.

[0434] When a user enters data, the system has the function of requesting, retrieving, and displaying past approval data and guideline data so that the user can refer to this data.

[0435] It has a function to set up a check execution button and a final application button, and to send data to the server according to the user's operation.

[0436] The emotion engine uses sensors such as cameras and microphones to analyze the user's facial expressions and tone of voice to estimate their emotional state. This information is sent to a server and used to adjust feedback.

[0437] 3. User:

[0438] Users use their devices to draft requests and revise them based on feedback sent back from the server.

[0439] Finally, the revised approval request is sent to the higher approver.

[0440] The device's camera and sensors provide input to the emotion engine, for example, entering approval details, running checks, and reviewing and correcting feedback.

[0441] Specific examples

[0442] For example, consider the case where a user submits a request for approval for a "new product development project."

[0443] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[0444] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0445] 3. The user uses the acquired data to fill out the details of the request (purpose, required budget, schedule, etc.).

[0446] 4. After filling out the form, the user clicks the "Check" button.

[0447] 5. The server analyzes the request and generates feedback saying "There is insufficient budget item details."

[0448] 6. The device's emotion engine analyzes the user's facial expressions and tone of voice to detect that the user is feeling slightly stressed.

[0449] 7. Based on this emotional information, the server generates more detailed suggestions for correction and polite feedback and sends it to the user.

[0450] 8. The user reviews the feedback and makes corrections, such as specifying budget items.

[0451] 9. Click the "Check" button again to request the server to check again.

[0452] 10. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[0453] 11. The server sends the request to the higher approver and the approval process begins.

[0454] In this way, the system takes into account the user's emotional state and realizes efficient and high-quality drafting of approval requests.

[0455] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0456] Step 1: Prepare the Database

[0457] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data. The input is the previously collected approval and guideline data, and the output is a database that can be efficiently searched.

[0458] Step 2: Initiate the drafting process

[0459] The user opens the approval proposal form on the terminal, selects the project category, and enters the title, purpose, content, budget, schedule, etc. of the approval. Using the user's selected category and detailed approval content information as input, the terminal sends this information to the server. As output, the request for approval content sent to the server is obtained.

[0460] Step 3: View past approvals and guidelines

[0461] Based on the received request, the server searches and retrieves data related to the target classification from the database. As input, it uses the user's submitted data and search conditions in the database, and as output, it obtains data on past approval requests and guideline data. It then sends this data to the terminal. The terminal displays this data so that the user can refer to it.

[0462] Step 4: Run automated checks

[0463] After the user has finished filling out the request, they click the "Check" button. The completed request is used as input, and the server receives this data and automatically analyzes it using a generative AI model. The output is a check of the accuracy and completeness of the request, resulting in feedback including corrections and points to check, which is sent to the device.

[0464] Step 5: Adjusting feedback

[0465] The device's emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice to estimate their emotional state. The user's facial expression and voice data are used as input, and the estimated emotional state is obtained as output. The server then adjusts the tone and content of the feedback based on this emotional state.

[0466] Step 6: Review and correct feedback

[0467] The user checks the feedback from the server on the terminal. The displayed feedback is used as input, and the user corrects the pointed out parts. After checking the corrections, the user clicks the "Check" button again to request a recheck from the server. As output, a request with the corrected approval content is sent to the server.

[0468] Step 7: Submit the final application

[0469] The server rechecks and confirms that all check points have been cleared. The revised approval request content is used as input, and feedback that it can be approved is sent to the terminal as output. The user clicks the "Final Application" button to send the final approval request content to the server. The server sends this final approval request content to the superior approver, and the approval process begins. The output is a status that the application to the superior approver has been completed.

[0470] (Application example 2)

[0471] 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."

[0472] Conventional approval proposal systems are inefficient in checking and correcting approval content, and do not provide feedback that takes into account the user's emotional state, which can lead to stress and reduced efficiency. Furthermore, similar problems can occur in the management of work instructions in factories and on-site, potentially resulting in reduced work efficiency and content quality.

[0473] 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.

[0474] In this invention, the server includes means for storing past approval request data and guideline data and making them efficiently searchable and accessible, means for displaying the past approval request data and guideline data so that they can be referenced on an input device used by a user to draft an approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to the approver, and means for recognizing the user's emotional state and adjusting the content and tone of the feedback. This enables flexible feedback that takes the user's emotional state into consideration, improving the efficiency and quality of approval request drafting and work instruction management.

[0475] The "server" is a device that stores past approval data and guideline data and makes them available for efficient search and reference.

[0476] The "database" is a system for efficiently managing, storing, and accessing past approval data and guideline data.

[0477] A "terminal" is an input device that allows a user to draft a request for approval and input and edit its contents.

[0478] The "emotion engine" is a system that analyzes a user's facial expressions and tone of voice to estimate their emotional state.

[0479] A "rei-i" is a document that records and processes proposals, applications, and approval processes within an organization.

[0480] The "check execution button" is an input means by which the user requests the server to automatically check the contents of the approval request.

[0481] The "final application button" is an input means for the user to send the final approval request content to the approver.

[0482] An "approver" is a person or system that has the authority to review the contents of the request and make final approval.

[0483] "Points to be corrected" are areas that require correction and are discovered when checking the contents of the request.

[0484] "Verification points" are areas that require verification and are discovered when checking the contents of the approval request.

[0485] "Classification" refers to the categories used to organize and index approval data and guideline data.

[0486] An "index" is an index that enables efficient reference to data.

[0487] To implement this invention, it is necessary to build a system that makes work instruction management more efficient and improves its quality through collaboration between the server, the terminal, and the user. This system includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[0488] System configuration

[0489] 1. Server:

[0490] The server stores past work instruction data and guideline data and manages the database so that they can be efficiently searched and referenced.

[0491] The server analyzes the contents of the work instructions sent by the user and automatically checks them based on guidelines and past data.

[0492] Check results and correction suggestions are generated and sent to the user's device.

[0493] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[0494] 2. Terminal:

[0495] The terminal provides an input form for the user to create a work instruction.

[0496] Users can view relevant historical and guideline data when creating work instructions.

[0497] It has a check execution button and a final send button, and has the function of sending the contents of the user's work instructions to the server.

[0498] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[0499] 3. User:

[0500] The user uses the terminal to create work instructions and revise the contents based on feedback from the server.

[0501] The final revised work instruction is sent to a higher approver.

[0502] The emotion engine receives input through the device's camera and sensors.

[0503] Program processing flow

[0504] The server stores past approval data and guideline data in a database, allowing users to efficiently refer to related data when drafting work instructions on their devices. Users enter the details of the work instructions through an input form, and the server performs automatic checks based on this. During this process, an emotion engine identifies the user's emotional state and generates appropriate feedback, which is sent to the device. This allows users to efficiently create and modify work instructions while receiving flexible support according to their emotional state.

[0505] Equipment and software used

[0506] Hardware: Camera (e.g. USB camera), device (e.g. tablet or PC)

[0507] Software: OpenCV (face detection), TENSORFLOW (registered trademark) (emotion recognition), custom feedback system

[0508] Specific examples

[0509] Consider the case of a factory line leader creating work instructions for a new production line. Using this system, the following steps are streamlined:

[0510] 1. The user creates a work instruction on the terminal.

[0511] 2. Send the work order to the server and perform an automatic check.

[0512] 3. The emotion engine analyzes the user's facial expressions and tone of voice and provides feedback according to their emotional state.

[0513] 4. The user modifies the work order based on the feedback and submits it to the server again.

[0514] 5. Click the final submit button to send the work order to the approver.

[0515] Prompt Sentence Examples

[0516] Enter the following information into your factory's work order creation support system:

[0517] 1. Work title

[0518] 2. Details of the work

[0519] 3. Materials needed

[0520] 4. Work Procedure

[0521] 5. Safety Guidelines

[0522] 6. User's real-time emotional state

[0523] This allows for flexible feedback that takes into account the user's emotional state, improving the efficiency and quality of work instruction management.

[0524] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0525] Step 1:

[0526] The server stores historical work instruction data and guideline data in a database, including work classification, project name, purpose, required materials, procedures, creators, etc., and indexes it to enable efficient searches. The input is historical work instruction data and guideline data, and the output is the indexed database.

[0527] Step 2:

[0528] The user opens the work instruction draft form on the terminal, selects the appropriate "classification," and begins to enter details such as the work title, purpose, content, required materials, procedures, schedule, etc. The input is the user's work instruction information, and the output is the draft work instruction that the user entered and saved on the terminal.

[0529] Step 3:

[0530] The device sends a request to the server based on the "Classification" field entered by the user. The server receives the request, retrieves data related to the target classification from the database, and returns it to the device. The input is the classification request made by the user, and the output is related historical data and guidelines.

[0531] Step 4:

[0532] After the user has finished writing the work instructions, he / she clicks the "Check" button. The terminal sends the work instructions to the server, which receives them. The input is the work instructions written by the user, and the output is the work instructions data received by the server.

[0533] Step 5:

[0534] The server analyzes the received work instruction content and automatically checks it. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The input is the work instruction content, and the output is the check results and correction suggestions based on specific criteria.

[0535] Step 6:

[0536] The emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. The device does this using cameras and sensors. The input is the user's real-time facial and voice data, and the output is the inferred emotional state.

[0537] Step 7:

[0538] The server adjusts the content and tone of the feedback based on information from the emotion engine. Depending on the emotional state, it refines the necessary correction suggestions and generates feedback that is considerate to the user. The input is the result of emotion recognition, and the output is the adjusted feedback content.

[0539] Step 8:

[0540] The user checks the feedback from the server on the terminal, confirms the points pointed out and points to note, and makes the necessary corrections. The input is the server's feedback, and the output is the corrected work instructions.

[0541] Step 9:

[0542] The user clicks the "Check" button again to request a recheck from the server. The server performs the recheck and confirms that all check points have been cleared. The input is the work instruction modified by the user and the recheck request, and the output is the result of the recheck.

[0543] Step 10:

[0544] After all items are cleared, the user clicks the "Final Submission" button and sends the final work order content to the server. The server sends the final work order content to the approver, and the approval process is then carried out. The input is the final modified work order, and the output is the work order sent to the approver.

[0545] 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.

[0546] 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 (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0547] 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.

[0548] [Second embodiment]

[0549] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0550] 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.

[0551] 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).

[0552] 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.

[0553] 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.

[0554] 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).

[0555] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0556] 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.

[0557] 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.

[0558] 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.

[0559] In the smart glasses 214, 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.

[0560] 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."

[0561] The embodiments of the present invention will be specifically described below.

[0562] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process. Below, we will explain the procedure for executing the program of this system and its operation with concrete examples.

[0563] System configuration

[0564] 1. Server:

[0565] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[0566] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[0567] Check results and correction suggestions are generated and sent to the user's device.

[0568] 2. Terminal:

[0569] The terminal provides a form for users to draft a request for approval.

[0570] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[0571] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[0572] 3. User:

[0573] Users use their devices to draft requests and revise them based on feedback from the server.

[0574] The final revised approval request is sent to the higher approver.

[0575] Program processing flow

[0576] 1. Prepare the database

[0577] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[0578] 2. Start of proposal process

[0579] The user opens the proposal form on their device and enters details such as the proposal title, purpose, content, budget, and schedule.

[0580] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[0581] 3. Display of past approval requests and guidelines

[0582] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends the data to the terminal.

[0583] The terminal displays this data for the user to refer to.

[0584] 4. Run an automatic check

[0585] After the user has finished writing the details of the request, they click the "Check" button on the terminal.

[0586] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[0587] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[0588] 5. Review and correct feedback

[0589] The user checks the feedback from the server on their device and makes corrections to the request based on the points pointed out and suggestions.

[0590] After making the corrections, click the "Check" button again to request the server to check again.

[0591] 6. Final application execution

[0592] The server will recheck to make sure all check items are cleared.

[0593] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[0594] The server sends the final approval to the higher approver, and the approval process begins.

[0595] Specific examples

[0596] As an example, consider the case where a user submits a request for approval for a "new product development project."

[0597] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[0598] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0599] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[0600] 4. After filling out the form, Tanaka clicks the "Check" button.

[0601] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[0602] 6. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[0603] 7. Click the "Check" button again to request a recheck of the server.

[0604] 8. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[0605] 9. The server sends the request to the higher approver and the approval process is completed.

[0606] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests.

[0607] The processing flow will be explained below.

[0608] Step 1:

[0609] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[0610] Step 2:

[0611] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[0612] Step 3:

[0613] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[0614] Step 4:

[0615] The server receives the request and retrieves data related to the target classification from a database and returns it.

[0616] Step 5:

[0617] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[0618] Step 6:

[0619] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[0620] Step 7:

[0621] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[0622] Step 8:

[0623] The server generates the analysis results and sends feedback to the device, including check points and suggested modifications.

[0624] Step 9:

[0625] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[0626] Step 10:

[0627] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[0628] Step 11:

[0629] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[0630] Step 12:

[0631] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[0632] Step 13:

[0633] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[0634] Example 1

[0635] 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."

[0636] Conventional approval drafting systems were unable to effectively utilize past approval data or guideline data, and the quality of automatic checks and feedback on approval content was limited, resulting in issues such as insufficient efficiency and quality improvement in approval drafting.In addition, there was insufficient support for users when revising approval content, and the final confirmation and higher-level approval process was often done manually, making the entire process time-consuming and labor-intensive.

[0637] 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.

[0638] In this invention, the server includes means for storing and searching past approval request data and guideline data, means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to a superior approver, means for analyzing the approval request content using an AI model and checking the accuracy and completeness of the content based on specific standards and rules, and means for providing the user with specific feedback based on the analysis results of the AI ​​model, thereby enabling the efficiency and quality of approval request drafting to be improved.

[0639] "Past approval data" refers to information on approval documents created in the past, including the approval classification, project name, purpose, budget, applicant, approval date, etc.

[0640] "Guideline data" is data that includes guidelines and regulations regarding the creation of approval documents established by companies and organizations, and describes specific procedures and standards.

[0641] "Terminal" refers to an electronic device that a user operates to draft a request for approval, including a PC, tablet, smartphone, etc.

[0642] A "server" refers to a computer system that manages a database and searches and processes data in response to user requests.

[0643] "Means of making something searchable" refers to the functionality for efficiently searching for information in a database and quickly obtaining the required data.

[0644] "Means for displaying" refers to the ability to visually present data on a terminal for a user to view.

[0645] "Means for automatic checking" refers to a function in which the system analyzes the content of the approval request entered by the user and verifies the accuracy and completeness of the content.

[0646] "Means for suggesting corrections and points to check" refers to a function that points out deficiencies and errors to the user based on the results of the automatic check and notifies them of points that need to be corrected.

[0647] "Means to check again" refers to a function that allows the user to re-analyze the contents of the revised approval request and reconfirm whether the confirmation items have been cleared.

[0648] "Superior approver" refers to an individual or departmental official who has the authority to approve the final contents of the request.

[0649] "Means of analyzing the contents of approval requests using AI models" refers to a function that uses artificial intelligence technologies such as machine learning and natural language processing to perform detailed analysis of approval requests.

[0650] "Means of providing specific feedback" refers to a function that presents specific suggestions for correction and points of attention to the user based on the results of the AI ​​model's analysis.

[0651] The present invention relates to a system for improving the efficiency and quality of approval drafting through collaboration between a server, a terminal, and a user. A method for specifically implementing the program of this system will be described below.

[0652] System Overview

[0653] This system involves collaboration between the server, terminals, and users to streamline and improve the quality of approval drafting. The server uses a database management system (DBMS) to manage past approval drafting data and guideline data, and analyzes the content of approval drafts sent by users. The analysis results are fed back to the user's terminal. The terminal provides an interface for users to draft approval drafts and displays data retrieved from the server. Users use their terminals to draft approval drafts and revise the content based on feedback from the server.

[0654] Hardware and Software Configuration

[0655] Server: The server manages the database using a DBMS such as MySQL or PostgreSQL, and uses a machine learning model (AI model) for analysis.

[0656] Device: Devices include PCs, tablets, smartphones, etc. The user interface is provided via a web browser or dedicated application.

[0657] User: The user operates the terminal to draft a request for approval and revise the content based on feedback from the server.

[0658] A concrete explanation of how the system works

[0659] 1. Prepare the database:

[0660] Server: The server stores past approval data and guideline data in a DBMS, including approval classification, project name, purpose, budget, applicant, approval date, etc. The data is indexed to enable efficient searches.

[0661] 2. Initiating a proposal:

[0662] User: The user opens the request draft form on their device and enters details such as the request title, purpose, content, budget, and schedule.

[0663] Terminal: The terminal sends a request to the server based on the "category" field entered by the user.

[0664] 3. View past approvals and guidelines:

[0665] Server: Receives the request, searches for past approval data and guideline data related to the target category, and sends it to the terminal. For example, execute the following query: "SELECT FROM Ringi WHERE category = 'New product development project'"

[0666] Terminal: The terminal displays this data for the user to refer to.

[0667] 4. Run the automatic check:

[0668] User: After filling out the details of the request, click the "Check" button on the device.

[0669] Server: The server receives the request and automatically analyzes it using AI models to check the accuracy and completeness of the content based on specific criteria and rules.

[0670] Example: "Please check if this request complies with the guidelines. Content: {Request content}"

[0671] Server: Generates feedback including points that need to be corrected and points to be noted, and sends this to the device.

[0672] 5. Review and correct feedback:

[0673] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[0674] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[0675] 6. Final application execution:

[0676] Server: Recheck and make sure all checks are cleared.

[0677] User: After receiving feedback from the server that "all items are acceptable," clicks the final submission button.

[0678] Server: The final approval request is sent to the higher approver, and the approval process begins.

[0679] Specific examples

[0680] As an example, consider the case of drafting a request for approval for a "new product development project."

[0681] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[0682] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0683] 3. The user uses this data to fill out the details of the request (purpose, required budget, schedule, etc.).

[0684] 4. After filling out the form, the user clicks the "Check" button.

[0685] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[0686] 6. The user checks the feedback and makes corrections, such as specifying budget items.

[0687] 7. Click the "Check" button again to request a recheck of the server.

[0688] 8. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[0689] 9. The server sends the request to the higher approver, and the approval process begins.

[0690] As described above, by implementing the system of the present invention, the efficiency and quality of approval drafting work can be improved.

[0691] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0692] Step 1:

[0693] Preparing the database

[0694] Server: A database management system (DBMS) is used to store past approval data and guideline data in a database. Specifically, DBMSs such as MySQL and PostgreSQL are used.

[0695] Input: Existing approval data and guideline data.

[0696] Data processing: Data indexing, classification, and import.

[0697] Output: A searchable database.

[0698] Specific operation: Executes SQL queries to store approval data and guideline data, and sets up indexes.

[0699] sql

[0700] CREATE INDEX idx_category ON Ringi(category);

[0701] Step 2:

[0702] Start of drafting a request

[0703] User: Uses a device to open the request draft form and enter details such as the request title, purpose, content, budget, and schedule.

[0704] Input: Title, purpose, content, budget, schedule, etc. of the request.

[0705] Output: A completed decision proposal form.

[0706] Specific actions: Enter data into an input form in a web browser or dedicated application. Input fields include, for example, "title," "purpose," and "content."

[0707] Step 3:

[0708] Display of past approval requests and guidelines

[0709] Terminal: Sends requests to the server based on the user's input data (especially the classification fields).

[0710] Input: The value of the classification field.

[0711] Output: Request information.

[0712] Specific operation: Encodes form data in JSON format and sends it to the server.

[0713] json

[0714] { "category": "New product development project"}

[0715] Server: Receives the request and searches the database for past approval requests and guideline data related to the target classification.

[0716] Input: Request information.

[0717] Data operations: Executing SQL queries.

[0718] Output: Past approval data and guideline data.

[0719] Specific behavior: Executes an SQL query and retrieves relevant data.

[0720] sql

[0721] SELECT FROM Ringi WHERE category = 'New product development project';

[0722] Terminal: Displays the retrieved data to the user.

[0723] Input: Data from the server.

[0724] Output: Past approvals and guidelines data displayed to the user.

[0725] Specific operation: Analyze the acquired data and display it in the user interface.

[0726] Step 4:

[0727] Running automatic checks

[0728] User: After filling out all the details of the request, click the "Check" button on the device.

[0729] Input: Completed request.

[0730] Output: The request to the server.

[0731] Specific behavior: Sends form data in JSON format to the server.

[0732] Server: Analyzes the received approval request using an AI model, checking the accuracy and completeness of the content based on specific criteria and rules.

[0733] Input: Data of the request.

[0734] Data calculation: Analysis using AI models.

[0735] Output: Feedback information.

[0736] Specific operations: Generate prompt sentences for the AI ​​model and perform analysis.

[0737] Example prompt: "Please check if this request complies with the guidelines. Content: {Request content}"

[0738] Server: Generates feedback including points that need to be corrected and points to be noted, and sends it to the device.

[0739] Input: Analysis results from the AI ​​model.

[0740] Data processing: Feedback generation.

[0741] Output: Feedback to the user.

[0742] Specific operation: Sends feedback data in JSON format to the device.

[0743] Step 5:

[0744] Review and correct feedback

[0745] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[0746] Input: Feedback from the server.

[0747] Output: Revised version of the approval request.

[0748] Specific actions: Rewrite and revise based on the feedback.

[0749] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[0750] Input: Revised version of the approval request.

[0751] Output: Re-request to the server.

[0752] Specific action: Send the corrected data to the server again.

[0753] Step 6:

[0754] Final application execution

[0755] Server: Recheck and make sure all checks are cleared.

[0756] Input: Revised version of the approval request.

[0757] Data calculation: Recheck process.

[0758] Output: Final feedback.

[0759] Specific actions: Use the AI ​​model again to confirm the content.

[0760] User: After receiving feedback from the server that "All items are acceptable," clicks the final submission button.

[0761] Input: Final feedback.

[0762] Output: Final application request.

[0763] Specific operation: Click the final application button to send it to the server.

[0764] Server: The final approval request is sent to the higher approver, and the approval process begins.

[0765] Input: Final approval content.

[0766] Data processing: Generate notifications to higher approvers.

[0767] Output: Notification to approver.

[0768] Specific behavior: Sends email or system notification to the senior approver and starts the approval process.

[0769] Through the above steps, the system of the present invention realizes efficient and quality improvement in drafting approval requests.

[0770] (Application example 1)

[0771] 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."

[0772] Currently, factories require a lot of time and effort to draft and review approval requests, resulting in problems such as reduced work efficiency and inconsistent quality. Furthermore, the review process, which involves referencing past approval data and guidelines, is often done manually, increasing the risk of human error. Furthermore, there is a need to automate these processes and significantly improve efficiency by incorporating the latest technology.

[0773] 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.

[0774] In this invention, the server includes a means for storing and searching past approval request data and guideline data, a means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, a means for automatically checking the content of the user's approval request and presenting corrections and confirmation points, a means for rechecking the approval request content revised by the user based on the corrections and confirmation points, a means for sending the final approval request content to a superior approver, a means for automatically drafting and confirming approval requests during the factory production process, a means for analyzing the content of the user's approval request using a generative AI model and generating correction proposals based on prompts, a terminal display means for confirming the generated correction proposals, and a means for automatically sending the approval request content to the approver after final confirmation, thereby enabling the efficiency and quality of approval request drafting and confirmation work to be improved.

[0775] "Past approval data" is a database that stores information about approvals that have been previously submitted and approved.

[0776] "Guideline data" is a database that stores information that indicates guidelines and standards for drafting approval requests.

[0777] "Means of making something searchable" refers to the ability to quickly find the desired information from a database.

[0778] "Proposal terminal" means the electronic device used by a user to create and edit a proposal.

[0779] "Means to automatically check and suggest corrections and points to check" refers to a function that analyzes the contents of the request based on an algorithm and notifies the user of problems and areas that need to be corrected.

[0780] "Means to check again" refers to a function that reanalyzes the contents of the approval request after the user has made corrections and checks whether the problem has been resolved.

[0781] "Means to send to a higher approver" refers to the function for transferring the contents of the request that ultimately require approval to the decision maker.

[0782] "Means for automatically drafting and confirming approval requests during the factory production process" refers to a function that automatically generates approval request content and confirms that content on the factory production line.

[0783] A "generative AI model" is an algorithmic model that uses machine learning to generate and analyze text.

[0784] The "means for generating correction suggestions based on prompt sentences" refers to a function that automatically presents optimal correction suggestions based on user input.

[0785] "Terminal display means" refers to a screen or monitor for displaying information referenced by the user.

[0786] As an embodiment of the present invention, the configuration of the following system and the processing of its programs will be specifically described. Below, we will explain a method for achieving efficiency and quality improvement in drafting and checking approval requests in the production process of a factory through cooperation between a server, a terminal, and a user.

[0787] System configuration

[0788] 1. Server:

[0789] The server stores past approval data and guideline data and manages the database so that they can be searched and referenced efficiently.

[0790] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[0791] A generative AI model is used to analyze the content of the user's request and generate suggested revisions based on the prompt text.

[0792] Check results and correction suggestions are generated and sent to the user's device.

[0793] 2. Terminal:

[0794] The terminal provides a form for users to draft a request for approval.

[0795] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[0796] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[0797] It has a display means for displaying the generated correction suggestions and points to note.

[0798] 3. User:

[0799] Users use their devices to draft requests and revise them based on feedback from the server.

[0800] The final revised approval request is sent to the higher approver.

[0801] Hardware and software used

[0802] Hardware

[0803] Factory robots

[0804] Smart devices (e.g., tablets)

[0805] server

[0806] software

[0807] Database Management Systems

[0808] Natural Language Processing Library

[0809] Generative AI models (e.g., ChatGPT)

[0810] Data processing and calculation

[0811] Database preparation

[0812] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[0813] Start of drafting a request

[0814] The user opens the approval proposal form on the terminal and enters details such as the title, purpose, content, budget, schedule, etc. The terminal then sends a request to the server based on the "category" field entered by the user when submitting the proposal.

[0815] Display of past approval requests and guidelines

[0816] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends it to the terminal. The terminal displays this data for the user to refer to.

[0817] Running automatic checks

[0818] After the user has finished filling out the request details, they click the "Check" button on their device. The server receives the request details and automatically analyzes them. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. A generative AI model is used to generate correction suggestions based on the prompt. As an example, the prompt used is, "I would like to create a request for approval regarding the introduction of a new production line. Please enter the purpose, required budget, and schedule."

[0819] Review and correct feedback

[0820] The user checks the feedback from the server on their device and corrects the approval request based on the points pointed out and suggestions. After correcting the corrections, they click the "Check" button again to request a recheck from the server.

[0821] Final application execution

[0822] The server rechecks and confirms that all check items have been cleared. The user receives feedback from the server that "all items are appropriable" and clicks the final application button. The server then sends the final approval request to the higher approver, and the approval process begins.

[0823] This will enable the efficiency of drafting and confirming approvals during the factory production process and improve quality.

[0824] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0825] Step 1:

[0826] The server stores and indexes past approval and guideline data in a database, making it possible to efficiently search the data. The server periodically updates this data to keep it up to date. The input is existing approval and guideline data, and the output is an indexed database.

[0827] Step 2:

[0828] The user opens the approval proposal form on the terminal and enters details such as the approval title, purpose, content, budget, schedule, etc. The terminal receives the user's input and sends the input data to the server. The input is the user's approval details, and the output is the data sent to the server.

[0829] Step 3:

[0830] The server receives the approval data entered by the user and searches the database for related past approval data and guideline data. The server sends the search results to the terminal and instructs it to display them. The input is the user's approval data and the database data, and the output is the search results for related past approval data and guideline data.

[0831] Step 4:

[0832] The terminal displays past approval data and guideline data received from the server for the user's reference. The user inputs and modifies the details of the approval while referring to the displayed data. The input is data from the server, and the output is displayed to the user.

[0833] Step 5:

[0834] After the user has finished entering the details of the approval request, they click the "Check" button on their device. The device then sends the user's approval request to the server. The input is the user's check request and the approval request details, and the output is the data sent to the server.

[0835] Step 6:

[0836] The server analyzes the received approval request and automatically checks it using a generative AI model. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The server generates correction suggestions based on the prompt text and sends the results to the terminal. The input is the user's approval request content and the generative AI model, and the output is correction suggestions and check results.

[0837] Step 7:

[0838] The terminal displays the correction suggestions and check results received from the server for the user to review. The user then corrects the approval request based on the feedback provided. The input is data from the server, and the output is the display and feedback to the user.

[0839] Step 8:

[0840] After the user has finished making the corrections, they click the "Check" button again, and the terminal sends the approval request content to the server again. The input is the approval request content after the user's corrections, and the output is the re-transmission to the server.

[0841] Step 9:

[0842] The server analyzes the approval request content again and checks whether all check items have been cleared. If the analysis results are satisfactory, it generates feedback stating "All items can be approved" and sends it to the terminal. The input is the revised approval request content and the generation AI model, and the output is the final feedback.

[0843] Step 10:

[0844] The user checks the final feedback and, if there are no problems, clicks the "Final Application" button. The terminal then sends the final approval request to the server. The input is the user's approval request and the final approval request, and the output is the final data sent to the server.

[0845] Step 11:

[0846] The server sends the received final approval content to the superior approver and starts the approval process. The input is the final approval content, and the output is sending the data to the superior approver and starting the approval process.

[0847] In this way, the process of drafting and reviewing approval requests is made more efficient, and quality is improved.

[0848] 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.

[0849] The embodiments of the present invention will be specifically described below.

[0850] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process.The present invention also includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[0851] System configuration

[0852] 1. Server:

[0853] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[0854] The server analyzes the content of the approval request sent by the user and automatically checks it based on guidelines and past approval request data.

[0855] Check results and correction suggestions are generated and sent to the user's device.

[0856] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[0857] 2. Terminal:

[0858] The terminal provides an input form for the user to draft a request for approval.

[0859] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[0860] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[0861] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[0862] 3. User:

[0863] Users use their devices to draft requests and revise them based on feedback from the server.

[0864] The final revised approval request is sent to the higher approver.

[0865] The emotion engine receives input through the device's camera and sensors.

[0866] Program processing flow

[0867] 1. Prepare the database

[0868] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes it for efficient searches.

[0869] 2. Start of proposal process

[0870] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[0871] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[0872] 3. Display of past approval requests and guidelines

[0873] The server receives the request and retrieves data related to the subject's classification from a database and returns it.

[0874] The terminal displays this data for the user to refer to.

[0875] 4. Run an automatic check

[0876] After the user has finished writing the details of the request, he / she clicks the "Check" button.

[0877] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[0878] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[0879] 5. Feedback adjustment

[0880] The device's emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. For example, if the user is feeling stressed, the device will provide more polite and detailed feedback.

[0881] The server adjusts the content and tone of the feedback based on information from the emotion engine.

[0882] 6. Reviewing and Correcting Feedback

[0883] The user checks the feedback from the server on the device, confirms the points pointed out and points to note, and makes any necessary corrections.

[0884] After making the corrections, click the "Check" button again to request the server to check again.

[0885] 7. Final application execution

[0886] The server will recheck to make sure all checkpoints are cleared.

[0887] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[0888] The server sends the final approval to the higher approver, and the approval process begins.

[0889] Specific examples

[0890] As an example, consider the case where a user submits a request for approval for a "new product development project."

[0891] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[0892] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0893] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[0894] 4. After filling out the form, Tanaka clicks the "Check" button.

[0895] 5. The server analyzes the request and returns feedback saying, "There is insufficient budget item detail."

[0896] 6. The device's emotion engine detects that Tanaka is feeling a little stressed.

[0897] 7. The server then generates more detailed suggestions for correction and feedback in a gentler tone and sends it to Tanaka.

[0898] 8. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[0899] 9. Click the "Check" button again to request a recheck of the server.

[0900] 10. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[0901] 11. The server sends the request to the higher approver and the approval process is completed.

[0902] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests, taking into account the emotional state of the user.

[0903] The processing flow will be explained below.

[0904] Step 1:

[0905] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[0906] Step 2:

[0907] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[0908] Step 3:

[0909] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[0910] Step 4:

[0911] The server receives the request and retrieves data related to the target classification from a database and returns it.

[0912] Step 5:

[0913] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[0914] Step 6:

[0915] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[0916] Step 7:

[0917] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[0918] Step 8:

[0919] To recognize the user's emotional state, the device uses cameras and sensors to analyze the user's facial expressions and tone of voice using an emotion engine.

[0920] Step 9:

[0921] The server generates the analysis results and adjusts the feedback based on the information from the emotion engine, for example, if the user is feeling stressed, the feedback will be presented in a gentler tone.

[0922] Step 10:

[0923] The server generates the check results and correction suggestions and sends them to the terminal.

[0924] Step 11:

[0925] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[0926] Step 12:

[0927] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[0928] Step 13:

[0929] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[0930] Step 14:

[0931] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[0932] Step 15:

[0933] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[0934] Example 2

[0935] 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."

[0936] In conventional approval proposal systems, users had to take a lot of time and effort to submit approvals, and the quality of the approvals was inconsistent. This could also cause stress and strain on users, making it difficult to create efficient, high-quality approval proposals. Furthermore, because feedback was uniform, flexible support that took into account the user's emotional state was not possible.

[0937] 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.

[0938] In this invention, the server includes a means for storing and retrieving past approval data and guideline data, a means for automatically checking the content of a user's approval and suggesting corrections and points to check, and a means for recognizing the user's emotional state and adjusting the method and content of feedback. This improves the efficiency of the user's approval drafting process and the quality of approval content. Furthermore, it is possible to reduce the user's stress and burden and provide flexible support according to the user's emotional state.

[0939] "Past approval data" refers to information related to approval documents that have previously been created or approved, which is stored in a database and is searchable.

[0940] "Guideline data" refers to documents and data that describe standard or recommended procedures, conditions, and standards for drafting approvals.

[0941] "Classification" refers to categories for organizing approval data and guideline data by purpose, project, or other criteria.

[0942] "Automatic checking" refers to the process in which the system analyzes the contents of a user's request and verifies it against pre-defined standards and rules.

[0943] "Points to correct and check" indicate the parts of the user's approval request that need to be corrected or items that should be checked, as identified as a result of the automated check.

[0944] "Emotional state" refers to the user's current psychological or emotional state, and is estimated from facial expressions, tone of voice, and the like.

[0945] "Feedback" refers to information and advice provided to the user by the system, which is useful for correcting and improving the content of the request.

[0946] "Superior approver" refers to the person who has the authority to make the final evaluation of the contents of the request and decide whether to approve or reject it.

[0947] "Indexing" refers to the process of assigning search keys to specific items so that information in a database can be efficiently searched.

[0948] A specific embodiment for carrying out the present invention will now be described. The system of the present invention works in cooperation with a server, a terminal, and a user to improve the efficiency and quality of drafting approval requests. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it also includes a function that provides flexible support.

[0949] System configuration

[0950] 1. Server:

[0951] The server manages a searchable database that stores past approval and guideline data, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data to enable efficient searches.

[0952] The server analyzes the content of the approval request submitted by the user and automatically checks it based on guidelines and past approval request data. Specifically, it uses a generative AI model to check the accuracy and completeness of the content.

[0953] The server generates the check results and correction suggestions and sends them to the user's device. This feedback is provided in a way that takes into account the user's emotional state. The content and tone of the feedback are adjusted based on information from the emotion engine.

[0954] Send the final approval request to a higher approver to begin the approval process.

[0955] 2. Terminal:

[0956] The terminal provides an input form for the user to submit a request for approval, which includes fields for entering details such as the title, purpose, content, budget, and schedule of the request.

[0957] When a user enters data, the system has the function of requesting, retrieving, and displaying past approval data and guideline data so that the user can refer to this data.

[0958] It has a function to set up a check execution button and a final application button, and to send data to the server according to the user's operation.

[0959] The emotion engine uses sensors such as cameras and microphones to analyze the user's facial expressions and tone of voice to estimate their emotional state. This information is sent to a server and used to adjust feedback.

[0960] 3. User:

[0961] Users use their devices to draft requests and revise them based on feedback sent back from the server.

[0962] Finally, the revised approval request is sent to the higher approver.

[0963] The device's camera and sensors provide input to the emotion engine, for example, entering approval details, running checks, and reviewing and correcting feedback.

[0964] Specific examples

[0965] For example, consider the case where a user submits a request for approval for a "new product development project."

[0966] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[0967] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[0968] 3. The user uses the acquired data to fill out the details of the request (purpose, required budget, schedule, etc.).

[0969] 4. After filling out the form, the user clicks the "Check" button.

[0970] 5. The server analyzes the request and generates feedback saying "There is insufficient budget item details."

[0971] 6. The device's emotion engine analyzes the user's facial expressions and tone of voice to detect that the user is feeling slightly stressed.

[0972] 7. Based on this emotional information, the server generates more detailed suggestions for correction and polite feedback and sends it to the user.

[0973] 8. The user reviews the feedback and makes corrections, such as specifying budget items.

[0974] 9. Click the "Check" button again to request the server to check again.

[0975] 10. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[0976] 11. The server sends the request to the higher approver and the approval process begins.

[0977] In this way, the system takes into account the user's emotional state and realizes efficient and high-quality drafting of approval requests.

[0978] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0979] Step 1: Prepare the Database

[0980] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data. The input is the previously collected approval and guideline data, and the output is a database that can be efficiently searched.

[0981] Step 2: Initiate the drafting process

[0982] The user opens the approval proposal form on the terminal, selects the project category, and enters the title, purpose, content, budget, schedule, etc. of the approval. Using the user's selected category and detailed approval content information as input, the terminal sends this information to the server. As output, the request for approval content sent to the server is obtained.

[0983] Step 3: View past approvals and guidelines

[0984] Based on the received request, the server searches and retrieves data related to the target classification from the database. As input, it uses the user's submitted data and search conditions in the database, and as output, it obtains data on past approval requests and guideline data. It then sends this data to the terminal. The terminal displays this data so that the user can refer to it.

[0985] Step 4: Run automated checks

[0986] After the user has finished filling out the request, they click the "Check" button. The completed request is used as input, and the server receives this data and automatically analyzes it using a generative AI model. The output is a check of the accuracy and completeness of the request, resulting in feedback including corrections and points to check, which is sent to the device.

[0987] Step 5: Adjusting feedback

[0988] The device's emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice to estimate their emotional state. The user's facial expression and voice data are used as input, and the estimated emotional state is obtained as output. The server then adjusts the tone and content of the feedback based on this emotional state.

[0989] Step 6: Review and correct feedback

[0990] The user checks the feedback from the server on the terminal. The displayed feedback is used as input, and the user corrects the pointed out parts. After checking the corrections, the user clicks the "Check" button again to request a recheck from the server. As output, a request with the corrected approval content is sent to the server.

[0991] Step 7: Submit the final application

[0992] The server rechecks and confirms that all check points have been cleared. The revised approval request content is used as input, and feedback that it can be approved is sent to the terminal as output. The user clicks the "Final Application" button to send the final approval request content to the server. The server sends this final approval request content to the superior approver, and the approval process begins. The output is a status that the application to the superior approver has been completed.

[0993] (Application example 2)

[0994] 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."

[0995] Conventional approval proposal systems are inefficient in checking and correcting approval content, and do not provide feedback that takes into account the user's emotional state, which can lead to stress and reduced efficiency. Furthermore, similar problems can occur in the management of work instructions in factories and on-site, potentially resulting in reduced work efficiency and content quality.

[0996] 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.

[0997] In this invention, the server includes means for storing past approval request data and guideline data and making them efficiently searchable and accessible, means for displaying the past approval request data and guideline data so that they can be referenced on an input device used by a user to draft an approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to the approver, and means for recognizing the user's emotional state and adjusting the content and tone of the feedback. This enables flexible feedback that takes the user's emotional state into consideration, improving the efficiency and quality of approval request drafting and work instruction management.

[0998] The "server" is a device that stores past approval data and guideline data and makes them available for efficient search and reference.

[0999] The "database" is a system for efficiently managing, storing, and accessing past approval data and guideline data.

[1000] A "terminal" is an input device that allows a user to draft a request for approval and input and edit its contents.

[1001] The "emotion engine" is a system that analyzes a user's facial expressions and tone of voice to estimate their emotional state.

[1002] A "rei-i" is a document that records and processes proposals, applications, and approval processes within an organization.

[1003] The "check execution button" is an input means by which the user requests the server to automatically check the contents of the approval request.

[1004] The "final application button" is an input means for the user to send the final approval request content to the approver.

[1005] An "approver" is a person or system that has the authority to review the contents of the request and make final approval.

[1006] "Points to be corrected" are areas that require correction and are discovered when checking the contents of the request.

[1007] "Verification points" are areas that require verification and are discovered when checking the contents of the approval request.

[1008] "Classification" refers to the categories used to organize and index approval data and guideline data.

[1009] An "index" is an index that enables efficient reference to data.

[1010] To implement this invention, it is necessary to build a system that makes work instruction management more efficient and improves its quality through collaboration between the server, the terminal, and the user. This system includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[1011] System configuration

[1012] 1. Server:

[1013] The server stores past work instruction data and guideline data and manages the database so that they can be efficiently searched and referenced.

[1014] The server analyzes the contents of the work instructions sent by the user and automatically checks them based on guidelines and past data.

[1015] Check results and correction suggestions are generated and sent to the user's device.

[1016] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[1017] 2. Terminal:

[1018] The terminal provides an input form for the user to create a work instruction.

[1019] Users can view relevant historical and guideline data when creating work instructions.

[1020] It has a check execution button and a final send button, and has the function of sending the contents of the user's work instructions to the server.

[1021] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[1022] 3. User:

[1023] The user uses the terminal to create work instructions and revise the contents based on feedback from the server.

[1024] The final revised work instruction is sent to a higher approver.

[1025] The emotion engine receives input through the device's camera and sensors.

[1026] Program processing flow

[1027] The server stores past approval data and guideline data in a database, allowing users to efficiently refer to related data when drafting work instructions on their devices. Users enter the details of the work instructions through an input form, and the server performs automatic checks based on this. During this process, an emotion engine identifies the user's emotional state and generates appropriate feedback, which is sent to the device. This allows users to efficiently create and modify work instructions while receiving flexible support according to their emotional state.

[1028] Equipment and software used

[1029] Hardware: Camera (e.g. USB camera), device (e.g. tablet or PC)

[1030] Software: OpenCV (face detection), TensorFlow (emotion recognition), custom feedback system

[1031] Specific examples

[1032] Consider the case of a factory line leader creating work instructions for a new production line. Using this system, the following steps are streamlined:

[1033] 1. The user creates a work instruction on the terminal.

[1034] 2. Send the work order to the server and perform an automatic check.

[1035] 3. The emotion engine analyzes the user's facial expressions and tone of voice and provides feedback according to their emotional state.

[1036] 4. The user modifies the work order based on the feedback and submits it to the server again.

[1037] 5. Click the final submit button to send the work order to the approver.

[1038] Prompt Sentence Examples

[1039] Enter the following information into your factory's work order creation support system:

[1040] 1. Work title

[1041] 2. Details of the work

[1042] 3. Materials needed

[1043] 4. Work Procedure

[1044] 5. Safety Guidelines

[1045] 6. User's real-time emotional state

[1046] This allows for flexible feedback that takes into account the user's emotional state, improving the efficiency and quality of work instruction management.

[1047] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1048] Step 1:

[1049] The server stores historical work instruction data and guideline data in a database, including work classification, project name, purpose, required materials, procedures, creators, etc., and indexes it to enable efficient searches. The input is historical work instruction data and guideline data, and the output is the indexed database.

[1050] Step 2:

[1051] The user opens the work instruction draft form on the terminal, selects the appropriate "classification," and begins to enter details such as the work title, purpose, content, required materials, procedures, schedule, etc. The input is the user's work instruction information, and the output is the draft work instruction that the user entered and saved on the terminal.

[1052] Step 3:

[1053] The device sends a request to the server based on the "Classification" field entered by the user. The server receives the request, retrieves data related to the target classification from the database, and returns it to the device. The input is the classification request made by the user, and the output is related historical data and guidelines.

[1054] Step 4:

[1055] After the user has finished writing the work instructions, he / she clicks the "Check" button. The terminal sends the work instructions to the server, which receives them. The input is the work instructions written by the user, and the output is the work instructions data received by the server.

[1056] Step 5:

[1057] The server analyzes the received work instruction content and automatically checks it. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The input is the work instruction content, and the output is the check results and correction suggestions based on specific criteria.

[1058] Step 6:

[1059] The emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. The device does this using cameras and sensors. The input is the user's real-time facial and voice data, and the output is the inferred emotional state.

[1060] Step 7:

[1061] The server adjusts the content and tone of the feedback based on information from the emotion engine. Depending on the emotional state, it refines the necessary correction suggestions and generates feedback that is considerate to the user. The input is the result of emotion recognition, and the output is the adjusted feedback content.

[1062] Step 8:

[1063] The user checks the feedback from the server on the terminal, confirms the points pointed out and points to note, and makes the necessary corrections. The input is the server's feedback, and the output is the corrected work instructions.

[1064] Step 9:

[1065] The user clicks the "Check" button again to request a recheck from the server. The server performs the recheck and confirms that all check points have been cleared. The input is the work instruction modified by the user and the recheck request, and the output is the result of the recheck.

[1066] Step 10:

[1067] After all items are cleared, the user clicks the "Final Submission" button and sends the final work order content to the server. The server sends the final work order content to the approver, and the approval process is then carried out. The input is the final modified work order, and the output is the work order sent to the approver.

[1068] 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.

[1069] 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.

[1070] 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.

[1071] [Third embodiment]

[1072] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1073] 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.

[1074] 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).

[1075] 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.

[1076] 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.

[1077] 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).

[1078] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1079] 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.

[1080] 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.

[1081] 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.

[1082] 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.

[1083] 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."

[1084] The embodiments of the present invention will be specifically described below.

[1085] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process. Below, we will explain the procedure for executing the program of this system and its operation with concrete examples.

[1086] System configuration

[1087] 1. Server:

[1088] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[1089] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[1090] Check results and correction suggestions are generated and sent to the user's device.

[1091] 2. Terminal:

[1092] The terminal provides a form for users to draft a request for approval.

[1093] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[1094] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[1095] 3. User:

[1096] Users use their devices to draft requests and revise them based on feedback from the server.

[1097] The final revised approval request is sent to the higher approver.

[1098] Program processing flow

[1099] 1. Prepare the database

[1100] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[1101] 2. Start of proposal process

[1102] The user opens the proposal form on their device and enters details such as the proposal title, purpose, content, budget, and schedule.

[1103] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[1104] 3. Display of past approval requests and guidelines

[1105] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends the data to the terminal.

[1106] The terminal displays this data for the user to refer to.

[1107] 4. Run an automatic check

[1108] After the user has finished writing the details of the request, they click the "Check" button on the terminal.

[1109] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[1110] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[1111] 5. Review and correct feedback

[1112] The user checks the feedback from the server on their device and makes corrections to the request based on the points pointed out and suggestions.

[1113] After making the corrections, click the "Check" button again to request the server to check again.

[1114] 6. Final application execution

[1115] The server will recheck to make sure all check items are cleared.

[1116] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[1117] The server sends the final approval to the higher approver, and the approval process begins.

[1118] Specific examples

[1119] As an example, consider the case where a user submits a request for approval for a "new product development project."

[1120] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[1121] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1122] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[1123] 4. After filling out the form, Tanaka clicks the "Check" button.

[1124] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[1125] 6. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[1126] 7. Click the "Check" button again to request a recheck of the server.

[1127] 8. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[1128] 9. The server sends the request to the higher approver and the approval process is completed.

[1129] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests.

[1130] The processing flow will be explained below.

[1131] Step 1:

[1132] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[1133] Step 2:

[1134] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[1135] Step 3:

[1136] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[1137] Step 4:

[1138] The server receives the request and retrieves data related to the target classification from a database and returns it.

[1139] Step 5:

[1140] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[1141] Step 6:

[1142] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[1143] Step 7:

[1144] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[1145] Step 8:

[1146] The server generates the analysis results and sends feedback to the device, including check points and suggested modifications.

[1147] Step 9:

[1148] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[1149] Step 10:

[1150] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[1151] Step 11:

[1152] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[1153] Step 12:

[1154] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[1155] Step 13:

[1156] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[1157] Example 1

[1158] 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."

[1159] Conventional approval drafting systems were unable to effectively utilize past approval data or guideline data, and the quality of automatic checks and feedback on approval content was limited, resulting in issues such as insufficient efficiency and quality improvement in approval drafting.In addition, there was insufficient support for users when revising approval content, and the final confirmation and higher-level approval process was often done manually, making the entire process time-consuming and labor-intensive.

[1160] 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.

[1161] In this invention, the server includes means for storing and searching past approval request data and guideline data, means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to a superior approver, means for analyzing the approval request content using an AI model and checking the accuracy and completeness of the content based on specific standards and rules, and means for providing the user with specific feedback based on the analysis results of the AI ​​model, thereby enabling the efficiency and quality of approval request drafting to be improved.

[1162] "Past approval data" refers to information on approval documents created in the past, including the approval classification, project name, purpose, budget, applicant, approval date, etc.

[1163] "Guideline data" is data that includes guidelines and regulations regarding the creation of approval documents established by companies and organizations, and describes specific procedures and standards.

[1164] "Terminal" refers to an electronic device that a user operates to draft a request for approval, including a PC, tablet, smartphone, etc.

[1165] A "server" refers to a computer system that manages a database and searches and processes data in response to user requests.

[1166] "Means of making something searchable" refers to the functionality for efficiently searching for information in a database and quickly obtaining the required data.

[1167] "Means for displaying" refers to the ability to visually present data on a terminal for a user to view.

[1168] "Means for automatic checking" refers to a function in which the system analyzes the content of the approval request entered by the user and verifies the accuracy and completeness of the content.

[1169] "Means for suggesting corrections and points to check" refers to a function that points out deficiencies and errors to the user based on the results of the automatic check and notifies them of points that need to be corrected.

[1170] "Means to check again" refers to a function that allows the user to re-analyze the contents of the revised approval request and reconfirm whether the confirmation items have been cleared.

[1171] "Superior approver" refers to an individual or departmental official who has the authority to approve the final contents of the request.

[1172] "Means of analyzing the contents of approval requests using AI models" refers to a function that uses artificial intelligence technologies such as machine learning and natural language processing to perform detailed analysis of approval requests.

[1173] "Means of providing specific feedback" refers to a function that presents specific suggestions for correction and points of attention to the user based on the results of the AI ​​model's analysis.

[1174] The present invention relates to a system for improving the efficiency and quality of approval drafting through collaboration between a server, a terminal, and a user. A method for specifically implementing the program of this system will be described below.

[1175] System Overview

[1176] This system involves collaboration between the server, terminals, and users to streamline and improve the quality of approval drafting. The server uses a database management system (DBMS) to manage past approval drafting data and guideline data, and analyzes the content of approval drafts sent by users. The analysis results are fed back to the user's terminal. The terminal provides an interface for users to draft approval drafts and displays data retrieved from the server. Users use their terminals to draft approval drafts and revise the content based on feedback from the server.

[1177] Hardware and Software Configuration

[1178] Server: The server manages the database using a DBMS such as MySQL or PostgreSQL, and uses a machine learning model (AI model) for analysis.

[1179] Device: Devices include PCs, tablets, smartphones, etc. The user interface is provided via a web browser or dedicated application.

[1180] User: The user operates the terminal to draft a request for approval and revise the content based on feedback from the server.

[1181] A concrete explanation of how the system works

[1182] 1. Prepare the database:

[1183] Server: The server stores past approval data and guideline data in a DBMS, including approval classification, project name, purpose, budget, applicant, approval date, etc. The data is indexed to enable efficient searches.

[1184] 2. Initiating a proposal:

[1185] User: The user opens the request draft form on their device and enters details such as the request title, purpose, content, budget, and schedule.

[1186] Terminal: The terminal sends a request to the server based on the "category" field entered by the user.

[1187] 3. View past approvals and guidelines:

[1188] Server: Receives the request, searches for past approval data and guideline data related to the target category, and sends it to the terminal. For example, execute the following query: "SELECT FROM Ringi WHERE category = 'New product development project'"

[1189] Terminal: The terminal displays this data for the user to refer to.

[1190] 4. Run the automatic check:

[1191] User: After filling out the details of the request, click the "Check" button on the device.

[1192] Server: The server receives the request and automatically analyzes it using AI models to check the accuracy and completeness of the content based on specific criteria and rules.

[1193] Example: "Please check if this request complies with the guidelines. Content: {Request content}"

[1194] Server: Generates feedback including points that need to be corrected and points to be noted, and sends this to the device.

[1195] 5. Review and correct feedback:

[1196] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[1197] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[1198] 6. Final application execution:

[1199] Server: Recheck and make sure all checks are cleared.

[1200] User: After receiving feedback from the server that "all items are acceptable," clicks the final submission button.

[1201] Server: The final approval request is sent to the higher approver, and the approval process begins.

[1202] Specific examples

[1203] As an example, consider the case of drafting a request for approval for a "new product development project."

[1204] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[1205] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1206] 3. The user uses this data to fill out the details of the request (purpose, required budget, schedule, etc.).

[1207] 4. After filling out the form, the user clicks the "Check" button.

[1208] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[1209] 6. The user checks the feedback and makes corrections, such as specifying budget items.

[1210] 7. Click the "Check" button again to request a recheck of the server.

[1211] 8. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[1212] 9. The server sends the request to the higher approver, and the approval process begins.

[1213] As described above, by implementing the system of the present invention, the efficiency and quality of approval drafting work can be improved.

[1214] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1215] Step 1:

[1216] Preparing the database

[1217] Server: A database management system (DBMS) is used to store past approval data and guideline data in a database. Specifically, DBMSs such as MySQL and PostgreSQL are used.

[1218] Input: Existing approval data and guideline data.

[1219] Data processing: Data indexing, classification, and import.

[1220] Output: A searchable database.

[1221] Specific operation: Executes SQL queries to store approval data and guideline data, and sets up indexes.

[1222] sql

[1223] CREATE INDEX idx_category ON Ringi(category);

[1224] Step 2:

[1225] Start of drafting a request

[1226] User: Uses a device to open the request draft form and enter details such as the request title, purpose, content, budget, and schedule.

[1227] Input: Title, purpose, content, budget, schedule, etc. of the request.

[1228] Output: A completed decision proposal form.

[1229] Specific actions: Enter data into an input form in a web browser or dedicated application. Input fields include, for example, "title," "purpose," and "content."

[1230] Step 3:

[1231] Display of past approval requests and guidelines

[1232] Terminal: Sends requests to the server based on the user's input data (especially the classification fields).

[1233] Input: The value of the classification field.

[1234] Output: Request information.

[1235] Specific operation: Encodes form data in JSON format and sends it to the server.

[1236] json

[1237] { "category": "New product development project"}

[1238] Server: Receives the request and searches the database for past approval requests and guideline data related to the target classification.

[1239] Input: Request information.

[1240] Data operations: Executing SQL queries.

[1241] Output: Past approval data and guideline data.

[1242] Specific behavior: Executes an SQL query and retrieves relevant data.

[1243] sql

[1244] SELECT FROM Ringi WHERE category = 'New product development project';

[1245] Terminal: Displays the retrieved data to the user.

[1246] Input: Data from the server.

[1247] Output: Past approvals and guidelines data displayed to the user.

[1248] Specific operation: Analyze the acquired data and display it in the user interface.

[1249] Step 4:

[1250] Running automatic checks

[1251] User: After filling out all the details of the request, click the "Check" button on the device.

[1252] Input: Completed request.

[1253] Output: The request to the server.

[1254] Specific behavior: Sends form data in JSON format to the server.

[1255] Server: Analyzes the received approval request using an AI model, checking the accuracy and completeness of the content based on specific criteria and rules.

[1256] Input: Data of the request.

[1257] Data calculation: Analysis using AI models.

[1258] Output: Feedback information.

[1259] Specific operations: Generate prompt sentences for the AI ​​model and perform analysis.

[1260] Example prompt: "Please check if this request complies with the guidelines. Content: {Request content}"

[1261] Server: Generates feedback including points that need to be corrected and points to be noted, and sends it to the device.

[1262] Input: Analysis results from the AI ​​model.

[1263] Data processing: Feedback generation.

[1264] Output: Feedback to the user.

[1265] Specific operation: Sends feedback data in JSON format to the device.

[1266] Step 5:

[1267] Review and correct feedback

[1268] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[1269] Input: Feedback from the server.

[1270] Output: Revised version of the approval request.

[1271] Specific actions: Rewrite and revise based on the feedback.

[1272] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[1273] Input: Revised version of the approval request.

[1274] Output: Re-request to the server.

[1275] Specific action: Send the corrected data to the server again.

[1276] Step 6:

[1277] Final application execution

[1278] Server: Recheck and make sure all checks are cleared.

[1279] Input: Revised version of the approval request.

[1280] Data calculation: Recheck process.

[1281] Output: Final feedback.

[1282] Specific actions: Use the AI ​​model again to confirm the content.

[1283] User: After receiving feedback from the server that "All items are acceptable," clicks the final submission button.

[1284] Input: Final feedback.

[1285] Output: Final application request.

[1286] Specific operation: Click the final application button to send it to the server.

[1287] Server: The final approval request is sent to the higher approver, and the approval process begins.

[1288] Input: Final approval content.

[1289] Data processing: Generate notifications to higher approvers.

[1290] Output: Notification to approver.

[1291] Specific behavior: Sends email or system notification to the senior approver and starts the approval process.

[1292] Through the above steps, the system of the present invention realizes efficient and quality improvement in drafting approval requests.

[1293] (Application example 1)

[1294] 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."

[1295] Currently, factories require a lot of time and effort to draft and review approval requests, resulting in problems such as reduced work efficiency and inconsistent quality. Furthermore, the review process, which involves referencing past approval data and guidelines, is often done manually, increasing the risk of human error. Furthermore, there is a need to automate these processes and significantly improve efficiency by incorporating the latest technology.

[1296] 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.

[1297] In this invention, the server includes a means for storing and searching past approval request data and guideline data, a means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, a means for automatically checking the content of the user's approval request and presenting corrections and confirmation points, a means for rechecking the approval request content revised by the user based on the corrections and confirmation points, a means for sending the final approval request content to a superior approver, a means for automatically drafting and confirming approval requests during the factory production process, a means for analyzing the content of the user's approval request using a generative AI model and generating correction proposals based on prompts, a terminal display means for confirming the generated correction proposals, and a means for automatically sending the approval request content to the approver after final confirmation, thereby enabling the efficiency and quality of approval request drafting and confirmation work to be improved.

[1298] "Past approval data" is a database that stores information about approvals that have been previously submitted and approved.

[1299] "Guideline data" is a database that stores information that indicates guidelines and standards for drafting approval requests.

[1300] "Means of making something searchable" refers to the ability to quickly find the desired information from a database.

[1301] "Proposal terminal" means the electronic device used by a user to create and edit a proposal.

[1302] "Means to automatically check and suggest corrections and points to check" refers to a function that analyzes the contents of the request based on an algorithm and notifies the user of problems and areas that need to be corrected.

[1303] "Means to check again" refers to a function that reanalyzes the contents of the approval request after the user has made corrections and checks whether the problem has been resolved.

[1304] "Means to send to a higher approver" refers to the function for transferring the contents of the request that ultimately require approval to the decision maker.

[1305] "Means for automatically drafting and confirming approval requests during the factory production process" refers to a function that automatically generates approval request content and confirms that content on the factory production line.

[1306] A "generative AI model" is an algorithmic model that uses machine learning to generate and analyze text.

[1307] The "means for generating correction suggestions based on prompt sentences" refers to a function that automatically presents optimal correction suggestions based on user input.

[1308] "Terminal display means" refers to a screen or monitor for displaying information referenced by the user.

[1309] As an embodiment of the present invention, the configuration of the following system and the processing of its programs will be specifically described. Below, we will explain a method for achieving efficiency and quality improvement in drafting and checking approval requests in the production process of a factory through cooperation between a server, a terminal, and a user.

[1310] System configuration

[1311] 1. Server:

[1312] The server stores past approval data and guideline data and manages the database so that they can be searched and referenced efficiently.

[1313] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[1314] A generative AI model is used to analyze the content of the user's request and generate suggested revisions based on the prompt text.

[1315] Check results and correction suggestions are generated and sent to the user's device.

[1316] 2. Terminal:

[1317] The terminal provides a form for users to draft a request for approval.

[1318] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[1319] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[1320] It has a display means for displaying the generated correction suggestions and points to note.

[1321] 3. User:

[1322] Users use their devices to draft requests and revise them based on feedback from the server.

[1323] The final revised approval request is sent to the higher approver.

[1324] Hardware and software used

[1325] Hardware

[1326] Factory robots

[1327] Smart devices (e.g., tablets)

[1328] server

[1329] software

[1330] Database Management Systems

[1331] Natural Language Processing Library

[1332] Generative AI models (e.g., ChatGPT)

[1333] Data processing and calculation

[1334] Database preparation

[1335] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[1336] Start of drafting a request

[1337] The user opens the approval proposal form on the terminal and enters details such as the title, purpose, content, budget, schedule, etc. The terminal then sends a request to the server based on the "category" field entered by the user when submitting the proposal.

[1338] Display of past approval requests and guidelines

[1339] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends it to the terminal. The terminal displays this data for the user to refer to.

[1340] Running automatic checks

[1341] After the user has finished filling out the request details, they click the "Check" button on their device. The server receives the request details and automatically analyzes them. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. A generative AI model is used to generate correction suggestions based on the prompt. As an example, the prompt used is, "I would like to create a request for approval regarding the introduction of a new production line. Please enter the purpose, required budget, and schedule."

[1342] Review and correct feedback

[1343] The user checks the feedback from the server on their device and corrects the approval request based on the points pointed out and suggestions. After correcting the corrections, they click the "Check" button again to request a recheck from the server.

[1344] Final application execution

[1345] The server rechecks and confirms that all check items have been cleared. The user receives feedback from the server that "all items are appropriable" and clicks the final application button. The server then sends the final approval request to the higher approver, and the approval process begins.

[1346] This will enable the efficiency of drafting and confirming approvals during the factory production process and improve quality.

[1347] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1348] Step 1:

[1349] The server stores and indexes past approval and guideline data in a database, making it possible to efficiently search the data. The server periodically updates this data to keep it up to date. The input is existing approval and guideline data, and the output is an indexed database.

[1350] Step 2:

[1351] The user opens the approval proposal form on the terminal and enters details such as the approval title, purpose, content, budget, schedule, etc. The terminal receives the user's input and sends the input data to the server. The input is the user's approval details, and the output is the data sent to the server.

[1352] Step 3:

[1353] The server receives the approval data entered by the user and searches the database for related past approval data and guideline data. The server sends the search results to the terminal and instructs it to display them. The input is the user's approval data and the database data, and the output is the search results for related past approval data and guideline data.

[1354] Step 4:

[1355] The terminal displays past approval data and guideline data received from the server for the user's reference. The user inputs and modifies the details of the approval while referring to the displayed data. The input is data from the server, and the output is displayed to the user.

[1356] Step 5:

[1357] After the user has finished entering the details of the approval request, they click the "Check" button on their device. The device then sends the user's approval request to the server. The input is the user's check request and the approval request details, and the output is the data sent to the server.

[1358] Step 6:

[1359] The server analyzes the received approval request and automatically checks it using a generative AI model. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The server generates correction suggestions based on the prompt text and sends the results to the terminal. The input is the user's approval request content and the generative AI model, and the output is correction suggestions and check results.

[1360] Step 7:

[1361] The terminal displays the correction suggestions and check results received from the server for the user to review. The user then corrects the approval request based on the feedback provided. The input is data from the server, and the output is the display and feedback to the user.

[1362] Step 8:

[1363] After the user has finished making the corrections, they click the "Check" button again, and the terminal sends the approval request content to the server again. The input is the approval request content after the user's corrections, and the output is the re-transmission to the server.

[1364] Step 9:

[1365] The server analyzes the approval request content again and checks whether all check items have been cleared. If the analysis results are satisfactory, it generates feedback stating "All items can be approved" and sends it to the terminal. The input is the revised approval request content and the generation AI model, and the output is the final feedback.

[1366] Step 10:

[1367] The user checks the final feedback and, if there are no problems, clicks the "Final Application" button. The terminal then sends the final approval request to the server. The input is the user's approval request and the final approval request, and the output is the final data sent to the server.

[1368] Step 11:

[1369] The server sends the received final approval content to the superior approver and starts the approval process. The input is the final approval content, and the output is sending the data to the superior approver and starting the approval process.

[1370] In this way, the process of drafting and reviewing approval requests is made more efficient, and quality is improved.

[1371] 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.

[1372] The embodiments of the present invention will be specifically described below.

[1373] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process.The present invention also includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[1374] System configuration

[1375] 1. Server:

[1376] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[1377] The server analyzes the content of the approval request sent by the user and automatically checks it based on guidelines and past approval request data.

[1378] Check results and correction suggestions are generated and sent to the user's device.

[1379] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[1380] 2. Terminal:

[1381] The terminal provides an input form for the user to draft a request for approval.

[1382] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[1383] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[1384] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[1385] 3. User:

[1386] Users use their devices to draft requests and revise them based on feedback from the server.

[1387] The final revised approval request is sent to the higher approver.

[1388] The emotion engine receives input through the device's camera and sensors.

[1389] Program processing flow

[1390] 1. Prepare the database

[1391] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes it for efficient searches.

[1392] 2. Start of proposal process

[1393] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[1394] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[1395] 3. Display of past approval requests and guidelines

[1396] The server receives the request and retrieves data related to the subject's classification from a database and returns it.

[1397] The terminal displays this data for the user to refer to.

[1398] 4. Run an automatic check

[1399] After the user has finished writing the details of the request, he / she clicks the "Check" button.

[1400] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[1401] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[1402] 5. Feedback adjustment

[1403] The device's emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. For example, if the user is feeling stressed, the device will provide more polite and detailed feedback.

[1404] The server adjusts the content and tone of the feedback based on information from the emotion engine.

[1405] 6. Reviewing and Correcting Feedback

[1406] The user checks the feedback from the server on the device, confirms the points pointed out and points to note, and makes any necessary corrections.

[1407] After making the corrections, click the "Check" button again to request the server to check again.

[1408] 7. Final application execution

[1409] The server will recheck to make sure all checkpoints are cleared.

[1410] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[1411] The server sends the final approval to the higher approver, and the approval process begins.

[1412] Specific examples

[1413] As an example, consider the case where a user submits a request for approval for a "new product development project."

[1414] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[1415] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1416] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[1417] 4. After filling out the form, Tanaka clicks the "Check" button.

[1418] 5. The server analyzes the request and returns feedback saying, "There is insufficient budget item detail."

[1419] 6. The device's emotion engine detects that Tanaka is feeling a little stressed.

[1420] 7. The server then generates more detailed suggestions for correction and feedback in a gentler tone and sends it to Tanaka.

[1421] 8. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[1422] 9. Click the "Check" button again to request a recheck of the server.

[1423] 10. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[1424] 11. The server sends the request to the higher approver and the approval process is completed.

[1425] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests, taking into account the emotional state of the user.

[1426] The processing flow will be explained below.

[1427] Step 1:

[1428] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[1429] Step 2:

[1430] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[1431] Step 3:

[1432] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[1433] Step 4:

[1434] The server receives the request and retrieves data related to the target classification from a database and returns it.

[1435] Step 5:

[1436] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[1437] Step 6:

[1438] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[1439] Step 7:

[1440] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[1441] Step 8:

[1442] To recognize the user's emotional state, the device uses cameras and sensors to analyze the user's facial expressions and tone of voice using an emotion engine.

[1443] Step 9:

[1444] The server generates the analysis results and adjusts the feedback based on the information from the emotion engine, for example, if the user is feeling stressed, the feedback will be presented in a gentler tone.

[1445] Step 10:

[1446] The server generates the check results and correction suggestions and sends them to the terminal.

[1447] Step 11:

[1448] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[1449] Step 12:

[1450] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[1451] Step 13:

[1452] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[1453] Step 14:

[1454] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[1455] Step 15:

[1456] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[1457] Example 2

[1458] 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."

[1459] In conventional approval proposal systems, users had to take a lot of time and effort to submit approvals, and the quality of the approvals was inconsistent. This could also cause stress and strain on users, making it difficult to create efficient, high-quality approval proposals. Furthermore, because feedback was uniform, flexible support that took into account the user's emotional state was not possible.

[1460] 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.

[1461] In this invention, the server includes a means for storing and retrieving past approval data and guideline data, a means for automatically checking the content of a user's approval and suggesting corrections and points to check, and a means for recognizing the user's emotional state and adjusting the method and content of feedback. This improves the efficiency of the user's approval drafting process and the quality of approval content. Furthermore, it is possible to reduce the user's stress and burden and provide flexible support according to the user's emotional state.

[1462] "Past approval data" refers to information related to approval documents that have previously been created or approved, which is stored in a database and is searchable.

[1463] "Guideline data" refers to documents and data that describe standard or recommended procedures, conditions, and standards for drafting approvals.

[1464] "Classification" refers to categories for organizing approval data and guideline data by purpose, project, or other criteria.

[1465] "Automatic checking" refers to the process in which the system analyzes the contents of a user's request and verifies it against pre-defined standards and rules.

[1466] "Points to correct and check" indicate the parts of the user's approval request that need to be corrected or items that should be checked, as identified as a result of the automated check.

[1467] "Emotional state" refers to the user's current psychological or emotional state, and is estimated from facial expressions, tone of voice, and the like.

[1468] "Feedback" refers to information and advice provided to the user by the system, which is useful for correcting and improving the content of the request.

[1469] "Superior approver" refers to the person who has the authority to make the final evaluation of the contents of the request and decide whether to approve or reject it.

[1470] "Indexing" refers to the process of assigning search keys to specific items so that information in a database can be efficiently searched.

[1471] A specific embodiment for carrying out the present invention will now be described. The system of the present invention works in cooperation with a server, a terminal, and a user to improve the efficiency and quality of drafting approval requests. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it also includes a function that provides flexible support.

[1472] System configuration

[1473] 1. Server:

[1474] The server manages a searchable database that stores past approval and guideline data, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data to enable efficient searches.

[1475] The server analyzes the content of the approval request submitted by the user and automatically checks it based on guidelines and past approval request data. Specifically, it uses a generative AI model to check the accuracy and completeness of the content.

[1476] The server generates the check results and correction suggestions and sends them to the user's device. This feedback is provided in a way that takes into account the user's emotional state. The content and tone of the feedback are adjusted based on information from the emotion engine.

[1477] Send the final approval request to a higher approver to begin the approval process.

[1478] 2. Terminal:

[1479] The terminal provides an input form for the user to submit a request for approval, which includes fields for entering details such as the title, purpose, content, budget, and schedule of the request.

[1480] When a user enters data, the system has the function of requesting, retrieving, and displaying past approval data and guideline data so that the user can refer to this data.

[1481] It has a function to set up a check execution button and a final application button, and to send data to the server according to the user's operation.

[1482] The emotion engine uses sensors such as cameras and microphones to analyze the user's facial expressions and tone of voice to estimate their emotional state. This information is sent to a server and used to adjust feedback.

[1483] 3. User:

[1484] Users use their devices to draft requests and revise them based on feedback sent back from the server.

[1485] Finally, the revised approval request is sent to the higher approver.

[1486] The device's camera and sensors provide input to the emotion engine, for example, entering approval details, running checks, and reviewing and correcting feedback.

[1487] Specific examples

[1488] For example, consider the case where a user submits a request for approval for a "new product development project."

[1489] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[1490] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1491] 3. The user uses the acquired data to fill out the details of the request (purpose, required budget, schedule, etc.).

[1492] 4. After filling out the form, the user clicks the "Check" button.

[1493] 5. The server analyzes the request and generates feedback saying "There is insufficient budget item details."

[1494] 6. The device's emotion engine analyzes the user's facial expressions and tone of voice to detect that the user is feeling slightly stressed.

[1495] 7. Based on this emotional information, the server generates more detailed suggestions for correction and polite feedback and sends it to the user.

[1496] 8. The user reviews the feedback and makes corrections, such as specifying budget items.

[1497] 9. Click the "Check" button again to request the server to check again.

[1498] 10. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[1499] 11. The server sends the request to the higher approver and the approval process begins.

[1500] In this way, the system takes into account the user's emotional state and realizes efficient and high-quality drafting of approval requests.

[1501] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1502] Step 1: Prepare the Database

[1503] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data. The input is the previously collected approval and guideline data, and the output is a database that can be efficiently searched.

[1504] Step 2: Initiate the drafting process

[1505] The user opens the approval proposal form on the terminal, selects the project category, and enters the title, purpose, content, budget, schedule, etc. of the approval. Using the user's selected category and detailed approval content information as input, the terminal sends this information to the server. As output, the request for approval content sent to the server is obtained.

[1506] Step 3: View past approvals and guidelines

[1507] Based on the received request, the server searches and retrieves data related to the target classification from the database. As input, it uses the user's submitted data and search conditions in the database, and as output, it obtains data on past approval requests and guideline data. It then sends this data to the terminal. The terminal displays this data so that the user can refer to it.

[1508] Step 4: Run automated checks

[1509] After the user has finished filling out the request, they click the "Check" button. The completed request is used as input, and the server receives this data and automatically analyzes it using a generative AI model. The output is a check of the accuracy and completeness of the request, resulting in feedback including corrections and points to check, which is sent to the device.

[1510] Step 5: Adjusting feedback

[1511] The device's emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice to estimate their emotional state. The user's facial expression and voice data are used as input, and the estimated emotional state is obtained as output. The server then adjusts the tone and content of the feedback based on this emotional state.

[1512] Step 6: Review and correct feedback

[1513] The user checks the feedback from the server on the terminal. The displayed feedback is used as input, and the user corrects the pointed out parts. After checking the corrections, the user clicks the "Check" button again to request a recheck from the server. As output, a request with the corrected approval content is sent to the server.

[1514] Step 7: Submit the final application

[1515] The server rechecks and confirms that all check points have been cleared. The revised approval request content is used as input, and feedback that it can be approved is sent to the terminal as output. The user clicks the "Final Application" button to send the final approval request content to the server. The server sends this final approval request content to the superior approver, and the approval process begins. The output is a status that the application to the superior approver has been completed.

[1516] (Application example 2)

[1517] 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."

[1518] Conventional approval proposal systems are inefficient in checking and correcting approval content, and do not provide feedback that takes into account the user's emotional state, which can lead to stress and reduced efficiency. Furthermore, similar problems can occur in the management of work instructions in factories and on-site, potentially resulting in reduced work efficiency and content quality.

[1519] 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.

[1520] In this invention, the server includes means for storing past approval request data and guideline data and making them efficiently searchable and accessible, means for displaying the past approval request data and guideline data so that they can be referenced on an input device used by a user to draft an approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to the approver, and means for recognizing the user's emotional state and adjusting the content and tone of the feedback. This enables flexible feedback that takes the user's emotional state into consideration, improving the efficiency and quality of approval request drafting and work instruction management.

[1521] The "server" is a device that stores past approval data and guideline data and makes them available for efficient search and reference.

[1522] The "database" is a system for efficiently managing, storing, and accessing past approval data and guideline data.

[1523] A "terminal" is an input device that allows a user to draft a request for approval and input and edit its contents.

[1524] The "emotion engine" is a system that analyzes a user's facial expressions and tone of voice to estimate their emotional state.

[1525] A "rei-i" is a document that records and processes proposals, applications, and approval processes within an organization.

[1526] The "check execution button" is an input means by which the user requests the server to automatically check the contents of the approval request.

[1527] The "final application button" is an input means for the user to send the final approval request content to the approver.

[1528] An "approver" is a person or system that has the authority to review the contents of the request and make final approval.

[1529] "Points to be corrected" are areas that require correction and are discovered when checking the contents of the request.

[1530] "Verification points" are areas that require verification and are discovered when checking the contents of the approval request.

[1531] "Classification" refers to the categories used to organize and index approval data and guideline data.

[1532] An "index" is an index that enables efficient reference to data.

[1533] To implement this invention, it is necessary to build a system that makes work instruction management more efficient and improves its quality through collaboration between the server, the terminal, and the user. This system includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[1534] System configuration

[1535] 1. Server:

[1536] The server stores past work instruction data and guideline data and manages the database so that they can be efficiently searched and referenced.

[1537] The server analyzes the contents of the work instructions sent by the user and automatically checks them based on guidelines and past data.

[1538] Check results and correction suggestions are generated and sent to the user's device.

[1539] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[1540] 2. Terminal:

[1541] The terminal provides an input form for the user to create a work instruction.

[1542] Users can view relevant historical and guideline data when creating work instructions.

[1543] It has a check execution button and a final send button, and has the function of sending the contents of the user's work instructions to the server.

[1544] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[1545] 3. User:

[1546] The user uses the terminal to create work instructions and revise the contents based on feedback from the server.

[1547] The final revised work instruction is sent to a higher approver.

[1548] The emotion engine receives input through the device's camera and sensors.

[1549] Program processing flow

[1550] The server stores past approval data and guideline data in a database, allowing users to efficiently refer to related data when drafting work instructions on their devices. Users enter the details of the work instructions through an input form, and the server performs automatic checks based on this. During this process, an emotion engine identifies the user's emotional state and generates appropriate feedback, which is sent to the device. This allows users to efficiently create and modify work instructions while receiving flexible support according to their emotional state.

[1551] Equipment and software used

[1552] Hardware: Camera (e.g. USB camera), device (e.g. tablet or PC)

[1553] Software: OpenCV (face detection), TensorFlow (emotion recognition), custom feedback system

[1554] Specific examples

[1555] Consider the case of a factory line leader creating work instructions for a new production line. Using this system, the following steps are streamlined:

[1556] 1. The user creates a work instruction on the terminal.

[1557] 2. Send the work order to the server and perform an automatic check.

[1558] 3. The emotion engine analyzes the user's facial expressions and tone of voice and provides feedback according to their emotional state.

[1559] 4. The user modifies the work order based on the feedback and submits it to the server again.

[1560] 5. Click the final submit button to send the work order to the approver.

[1561] Prompt Sentence Examples

[1562] Enter the following information into your factory's work order creation support system:

[1563] 1. Work title

[1564] 2. Details of the work

[1565] 3. Materials needed

[1566] 4. Work Procedure

[1567] 5. Safety Guidelines

[1568] 6. User's real-time emotional state

[1569] This allows for flexible feedback that takes into account the user's emotional state, improving the efficiency and quality of work instruction management.

[1570] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1571] Step 1:

[1572] The server stores historical work instruction data and guideline data in a database, including work classification, project name, purpose, required materials, procedures, creators, etc., and indexes it to enable efficient searches. The input is historical work instruction data and guideline data, and the output is the indexed database.

[1573] Step 2:

[1574] The user opens the work instruction draft form on the terminal, selects the appropriate "classification," and begins to enter details such as the work title, purpose, content, required materials, procedures, schedule, etc. The input is the user's work instruction information, and the output is the draft work instruction that the user entered and saved on the terminal.

[1575] Step 3:

[1576] The device sends a request to the server based on the "Classification" field entered by the user. The server receives the request, retrieves data related to the target classification from the database, and returns it to the device. The input is the classification request made by the user, and the output is related historical data and guidelines.

[1577] Step 4:

[1578] After the user has finished writing the work instructions, he / she clicks the "Check" button. The terminal sends the work instructions to the server, which receives them. The input is the work instructions written by the user, and the output is the work instructions data received by the server.

[1579] Step 5:

[1580] The server analyzes the received work instruction content and automatically checks it. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The input is the work instruction content, and the output is the check results and correction suggestions based on specific criteria.

[1581] Step 6:

[1582] The emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. The device does this using cameras and sensors. The input is the user's real-time facial and voice data, and the output is the inferred emotional state.

[1583] Step 7:

[1584] The server adjusts the content and tone of the feedback based on information from the emotion engine. Depending on the emotional state, it refines the necessary correction suggestions and generates feedback that is considerate to the user. The input is the result of emotion recognition, and the output is the adjusted feedback content.

[1585] Step 8:

[1586] The user checks the feedback from the server on the terminal, confirms the points pointed out and points to note, and makes the necessary corrections. The input is the server's feedback, and the output is the corrected work instructions.

[1587] Step 9:

[1588] The user clicks the "Check" button again to request a recheck from the server. The server performs the recheck and confirms that all check points have been cleared. The input is the work instruction modified by the user and the recheck request, and the output is the result of the recheck.

[1589] Step 10:

[1590] After all items are cleared, the user clicks the "Final Submission" button and sends the final work order content to the server. The server sends the final work order content to the approver, and the approval process is then carried out. The input is the final modified work order, and the output is the work order sent to the approver.

[1591] 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.

[1592] 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.

[1593] 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.

[1594] [Fourth embodiment]

[1595] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1596] 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.

[1597] 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).

[1598] 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.

[1599] 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.

[1600] 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).

[1601] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1602] 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.

[1603] 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.

[1604] 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.

[1605] 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.

[1606] 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.

[1607] 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."

[1608] The embodiments of the present invention will be specifically described below.

[1609] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process. Below, we will explain the procedure for executing the program of this system and its operation with concrete examples.

[1610] System configuration

[1611] 1. Server:

[1612] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[1613] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[1614] Check results and correction suggestions are generated and sent to the user's device.

[1615] 2. Terminal:

[1616] The terminal provides a form for users to draft a request for approval.

[1617] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[1618] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[1619] 3. User:

[1620] Users use their devices to draft requests and revise them based on feedback from the server.

[1621] The final revised approval request is sent to the higher approver.

[1622] Program processing flow

[1623] 1. Prepare the database

[1624] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[1625] 2. Start of proposal process

[1626] The user opens the proposal form on their device and enters details such as the proposal title, purpose, content, budget, and schedule.

[1627] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[1628] 3. Display of past approval requests and guidelines

[1629] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends the data to the terminal.

[1630] The terminal displays this data for the user to refer to.

[1631] 4. Run an automatic check

[1632] After the user has finished writing the details of the request, they click the "Check" button on the terminal.

[1633] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[1634] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[1635] 5. Review and correct feedback

[1636] The user checks the feedback from the server on their device and makes corrections to the request based on the points pointed out and suggestions.

[1637] After making the corrections, click the "Check" button again to request the server to check again.

[1638] 6. Final application execution

[1639] The server will recheck to make sure all check items are cleared.

[1640] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[1641] The server sends the final approval to the higher approver, and the approval process begins.

[1642] Specific examples

[1643] As an example, consider the case where a user submits a request for approval for a "new product development project."

[1644] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[1645] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1646] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[1647] 4. After filling out the form, Tanaka clicks the "Check" button.

[1648] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[1649] 6. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[1650] 7. Click the "Check" button again to request a recheck of the server.

[1651] 8. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[1652] 9. The server sends the request to the higher approver and the approval process is completed.

[1653] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests.

[1654] The processing flow will be explained below.

[1655] Step 1:

[1656] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[1657] Step 2:

[1658] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[1659] Step 3:

[1660] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[1661] Step 4:

[1662] The server receives the request and retrieves data related to the target classification from a database and returns it.

[1663] Step 5:

[1664] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[1665] Step 6:

[1666] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[1667] Step 7:

[1668] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[1669] Step 8:

[1670] The server generates the analysis results and sends feedback to the device, including check points and suggested modifications.

[1671] Step 9:

[1672] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[1673] Step 10:

[1674] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[1675] Step 11:

[1676] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[1677] Step 12:

[1678] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[1679] Step 13:

[1680] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[1681] Example 1

[1682] 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."

[1683] Conventional approval drafting systems were unable to effectively utilize past approval data or guideline data, and the quality of automatic checks and feedback on approval content was limited, resulting in issues such as insufficient efficiency and quality improvement in approval drafting.In addition, there was insufficient support for users when revising approval content, and the final confirmation and higher-level approval process was often done manually, making the entire process time-consuming and labor-intensive.

[1684] 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.

[1685] In this invention, the server includes means for storing and searching past approval request data and guideline data, means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to a superior approver, means for analyzing the approval request content using an AI model and checking the accuracy and completeness of the content based on specific standards and rules, and means for providing the user with specific feedback based on the analysis results of the AI ​​model, thereby enabling the efficiency and quality of approval request drafting to be improved.

[1686] "Past approval data" refers to information on approval documents created in the past, including the approval classification, project name, purpose, budget, applicant, approval date, etc.

[1687] "Guideline data" is data that includes guidelines and regulations regarding the creation of approval documents established by companies and organizations, and describes specific procedures and standards.

[1688] "Terminal" refers to an electronic device that a user operates to draft a request for approval, including a PC, tablet, smartphone, etc.

[1689] A "server" refers to a computer system that manages a database and searches and processes data in response to user requests.

[1690] "Means of making something searchable" refers to the functionality for efficiently searching for information in a database and quickly obtaining the required data.

[1691] "Means for displaying" refers to the ability to visually present data on a terminal for a user to view.

[1692] "Means for automatic checking" refers to a function in which the system analyzes the content of the approval request entered by the user and verifies the accuracy and completeness of the content.

[1693] "Means for suggesting corrections and points to check" refers to a function that points out deficiencies and errors to the user based on the results of the automatic check and notifies them of points that need to be corrected.

[1694] "Means to check again" refers to a function that allows the user to re-analyze the contents of the revised approval request and reconfirm whether the confirmation items have been cleared.

[1695] "Superior approver" refers to an individual or departmental official who has the authority to approve the final contents of the request.

[1696] "Means of analyzing the contents of approval requests using AI models" refers to a function that uses artificial intelligence technologies such as machine learning and natural language processing to perform detailed analysis of approval requests.

[1697] "Means of providing specific feedback" refers to a function that presents specific suggestions for correction and points of attention to the user based on the results of the AI ​​model's analysis.

[1698] The present invention relates to a system for improving the efficiency and quality of approval drafting through collaboration between a server, a terminal, and a user. A method for specifically implementing the program of this system will be described below.

[1699] System Overview

[1700] This system involves collaboration between the server, terminals, and users to streamline and improve the quality of approval drafting. The server uses a database management system (DBMS) to manage past approval drafting data and guideline data, and analyzes the content of approval drafts sent by users. The analysis results are fed back to the user's terminal. The terminal provides an interface for users to draft approval drafts and displays data retrieved from the server. Users use their terminals to draft approval drafts and revise the content based on feedback from the server.

[1701] Hardware and Software Configuration

[1702] Server: The server manages the database using a DBMS such as MySQL or PostgreSQL, and uses a machine learning model (AI model) for analysis.

[1703] Device: Devices include PCs, tablets, smartphones, etc. The user interface is provided via a web browser or dedicated application.

[1704] User: The user operates the terminal to draft a request for approval and revise the content based on feedback from the server.

[1705] A concrete explanation of how the system works

[1706] 1. Prepare the database:

[1707] Server: The server stores past approval data and guideline data in a DBMS, including approval classification, project name, purpose, budget, applicant, approval date, etc. The data is indexed to enable efficient searches.

[1708] 2. Initiating a proposal:

[1709] User: The user opens the request draft form on their device and enters details such as the request title, purpose, content, budget, and schedule.

[1710] Terminal: The terminal sends a request to the server based on the "category" field entered by the user.

[1711] 3. View past approvals and guidelines:

[1712] Server: Receives the request, searches for past approval data and guideline data related to the target category, and sends it to the terminal. For example, execute the following query: "SELECT FROM Ringi WHERE category = 'New product development project'"

[1713] Terminal: The terminal displays this data for the user to refer to.

[1714] 4. Run the automatic check:

[1715] User: After filling out the details of the request, click the "Check" button on the device.

[1716] Server: The server receives the request and automatically analyzes it using AI models to check the accuracy and completeness of the content based on specific criteria and rules.

[1717] Example: "Please check if this request complies with the guidelines. Content: {Request content}"

[1718] Server: Generates feedback including points that need to be corrected and points to be noted, and sends this to the device.

[1719] 5. Review and correct feedback:

[1720] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[1721] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[1722] 6. Final application execution:

[1723] Server: Recheck and make sure all checks are cleared.

[1724] User: After receiving feedback from the server that "all items are acceptable," clicks the final submission button.

[1725] Server: The final approval request is sent to the higher approver, and the approval process begins.

[1726] Specific examples

[1727] As an example, consider the case of drafting a request for approval for a "new product development project."

[1728] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[1729] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1730] 3. The user uses this data to fill out the details of the request (purpose, required budget, schedule, etc.).

[1731] 4. After filling out the form, the user clicks the "Check" button.

[1732] 5. The server analyzes the request and returns feedback, for example, "There is insufficient budget item detail."

[1733] 6. The user checks the feedback and makes corrections, such as specifying budget items.

[1734] 7. Click the "Check" button again to request a recheck of the server.

[1735] 8. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[1736] 9. The server sends the request to the higher approver, and the approval process begins.

[1737] As described above, by implementing the system of the present invention, the efficiency and quality of approval drafting work can be improved.

[1738] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1739] Step 1:

[1740] Preparing the database

[1741] Server: A database management system (DBMS) is used to store past approval data and guideline data in a database. Specifically, DBMSs such as MySQL and PostgreSQL are used.

[1742] Input: Existing approval data and guideline data.

[1743] Data processing: Data indexing, classification, and import.

[1744] Output: A searchable database.

[1745] Specific operation: Executes SQL queries to store approval data and guideline data, and sets up indexes.

[1746] sql

[1747] CREATE INDEX idx_category ON Ringi(category);

[1748] Step 2:

[1749] Start of drafting a request

[1750] User: Uses a device to open the request draft form and enter details such as the request title, purpose, content, budget, and schedule.

[1751] Input: Title, purpose, content, budget, schedule, etc. of the request.

[1752] Output: A completed decision proposal form.

[1753] Specific actions: Enter data into an input form in a web browser or dedicated application. Input fields include, for example, "title," "purpose," and "content."

[1754] Step 3:

[1755] Display of past approval requests and guidelines

[1756] Terminal: Sends requests to the server based on the user's input data (especially the classification fields).

[1757] Input: The value of the classification field.

[1758] Output: Request information.

[1759] Specific operation: Encodes form data in JSON format and sends it to the server.

[1760] json

[1761] { "category": "New product development project"}

[1762] Server: Receives the request and searches the database for past approval requests and guideline data related to the target classification.

[1763] Input: Request information.

[1764] Data operations: Executing SQL queries.

[1765] Output: Past approval data and guideline data.

[1766] Specific behavior: Executes an SQL query and retrieves relevant data.

[1767] sql

[1768] SELECT FROM Ringi WHERE category = 'New product development project';

[1769] Terminal: Displays the retrieved data to the user.

[1770] Input: Data from the server.

[1771] Output: Past approvals and guidelines data displayed to the user.

[1772] Specific operation: Analyze the acquired data and display it in the user interface.

[1773] Step 4:

[1774] Running automatic checks

[1775] User: After filling out all the details of the request, click the "Check" button on the device.

[1776] Input: Completed request.

[1777] Output: The request to the server.

[1778] Specific behavior: Sends form data in JSON format to the server.

[1779] Server: Analyzes the received approval request using an AI model, checking the accuracy and completeness of the content based on specific criteria and rules.

[1780] Input: Data of the request.

[1781] Data calculation: Analysis using AI models.

[1782] Output: Feedback information.

[1783] Specific operations: Generate prompt sentences for the AI ​​model and perform analysis.

[1784] Example prompt: "Please check if this request complies with the guidelines. Content: {Request content}"

[1785] Server: Generates feedback including points that need to be corrected and points to be noted, and sends it to the device.

[1786] Input: Analysis results from the AI ​​model.

[1787] Data processing: Feedback generation.

[1788] Output: Feedback to the user.

[1789] Specific operation: Sends feedback data in JSON format to the device.

[1790] Step 5:

[1791] Review and correct feedback

[1792] User: Check the feedback from the server and revise the request based on the issues and suggestions.

[1793] Input: Feedback from the server.

[1794] Output: Revised version of the approval request.

[1795] Specific actions: Rewrite and revise based on the feedback.

[1796] User: After making the corrections, click the "Check" button again to request a recheck from the server.

[1797] Input: Revised version of the approval request.

[1798] Output: Re-request to the server.

[1799] Specific action: Send the corrected data to the server again.

[1800] Step 6:

[1801] Final application execution

[1802] Server: Recheck and make sure all checks are cleared.

[1803] Input: Revised version of the approval request.

[1804] Data calculation: Recheck process.

[1805] Output: Final feedback.

[1806] Specific actions: Use the AI ​​model again to confirm the content.

[1807] User: After receiving feedback from the server that "All items are acceptable," clicks the final submission button.

[1808] Input: Final feedback.

[1809] Output: Final application request.

[1810] Specific operation: Click the final application button to send it to the server.

[1811] Server: The final approval request is sent to the higher approver, and the approval process begins.

[1812] Input: Final approval content.

[1813] Data processing: Generate notifications to higher approvers.

[1814] Output: Notification to approver.

[1815] Specific behavior: Sends email or system notification to the senior approver and starts the approval process.

[1816] Through the above steps, the system of the present invention realizes efficient and quality improvement in drafting approval requests.

[1817] (Application example 1)

[1818] 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."

[1819] Currently, factories require a lot of time and effort to draft and review approval requests, resulting in problems such as reduced work efficiency and inconsistent quality. Furthermore, the review process, which involves referencing past approval data and guidelines, is often done manually, increasing the risk of human error. Furthermore, there is a need to automate these processes and significantly improve efficiency by incorporating the latest technology.

[1820] 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.

[1821] In this invention, the server includes a means for storing and searching past approval request data and guideline data, a means for displaying the past approval request data and guideline data so that they can be referenced on the terminal where the user drafts the approval request, a means for automatically checking the content of the user's approval request and presenting corrections and confirmation points, a means for rechecking the approval request content revised by the user based on the corrections and confirmation points, a means for sending the final approval request content to a superior approver, a means for automatically drafting and confirming approval requests during the factory production process, a means for analyzing the content of the user's approval request using a generative AI model and generating correction proposals based on prompts, a terminal display means for confirming the generated correction proposals, and a means for automatically sending the approval request content to the approver after final confirmation, thereby enabling the efficiency and quality of approval request drafting and confirmation work to be improved.

[1822] "Past approval data" is a database that stores information about approvals that have been previously submitted and approved.

[1823] "Guideline data" is a database that stores information that indicates guidelines and standards for drafting approval requests.

[1824] "Means of making something searchable" refers to the ability to quickly find the desired information from a database.

[1825] "Proposal terminal" means the electronic device used by a user to create and edit a proposal.

[1826] "Means to automatically check and suggest corrections and points to check" refers to a function that analyzes the contents of the request based on an algorithm and notifies the user of problems and areas that need to be corrected.

[1827] "Means to check again" refers to a function that reanalyzes the contents of the approval request after the user has made corrections and checks whether the problem has been resolved.

[1828] "Means to send to a higher approver" refers to the function for transferring the contents of the request that ultimately require approval to the decision maker.

[1829] "Means for automatically drafting and confirming approval requests during the factory production process" refers to a function that automatically generates approval request content and confirms that content on the factory production line.

[1830] A "generative AI model" is an algorithmic model that uses machine learning to generate and analyze text.

[1831] The "means for generating correction suggestions based on prompt sentences" refers to a function that automatically presents optimal correction suggestions based on user input.

[1832] "Terminal display means" refers to a screen or monitor for displaying information referenced by the user.

[1833] As an embodiment of the present invention, the configuration of the following system and the processing of its programs will be specifically described. Below, we will explain a method for achieving efficiency and quality improvement in drafting and checking approval requests in the production process of a factory through cooperation between a server, a terminal, and a user.

[1834] System configuration

[1835] 1. Server:

[1836] The server stores past approval data and guideline data and manages the database so that they can be searched and referenced efficiently.

[1837] The server analyzes the content of the request sent by the user and checks it based on guidelines and past request data.

[1838] A generative AI model is used to analyze the content of the user's request and generate suggested revisions based on the prompt text.

[1839] Check results and correction suggestions are generated and sent to the user's device.

[1840] 2. Terminal:

[1841] The terminal provides a form for users to draft a request for approval.

[1842] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[1843] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[1844] It has a display means for displaying the generated correction suggestions and points to note.

[1845] 3. User:

[1846] Users use their devices to draft requests and revise them based on feedback from the server.

[1847] The final revised approval request is sent to the higher approver.

[1848] Hardware and software used

[1849] Hardware

[1850] Factory robots

[1851] Smart devices (e.g., tablets)

[1852] server

[1853] software

[1854] Database Management Systems

[1855] Natural Language Processing Library

[1856] Generative AI models (e.g., ChatGPT)

[1857] Data processing and calculation

[1858] Database preparation

[1859] The server stores past approval data and guideline data in a database. This data includes approval classification, project name, purpose, budget, applicant, approval date, etc., and is indexed to enable efficient searches.

[1860] Start of drafting a request

[1861] The user opens the approval proposal form on the terminal and enters details such as the title, purpose, content, budget, schedule, etc. The terminal then sends a request to the server based on the "category" field entered by the user when submitting the proposal.

[1862] Display of past approval requests and guidelines

[1863] The server receives the request, searches for past approval requests and guideline data related to the target classification, and sends it to the terminal. The terminal displays this data for the user to refer to.

[1864] Running automatic checks

[1865] After the user has finished filling out the request details, they click the "Check" button on their device. The server receives the request details and automatically analyzes them. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. A generative AI model is used to generate correction suggestions based on the prompt. As an example, the prompt used is, "I would like to create a request for approval regarding the introduction of a new production line. Please enter the purpose, required budget, and schedule."

[1866] Review and correct feedback

[1867] The user checks the feedback from the server on their device and corrects the approval request based on the points pointed out and suggestions. After correcting the corrections, they click the "Check" button again to request a recheck from the server.

[1868] Final application execution

[1869] The server rechecks and confirms that all check items have been cleared. The user receives feedback from the server that "all items are appropriable" and clicks the final application button. The server then sends the final approval request to the higher approver, and the approval process begins.

[1870] This will enable the efficiency of drafting and confirming approvals during the factory production process and improve quality.

[1871] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1872] Step 1:

[1873] The server stores and indexes past approval and guideline data in a database, making it possible to efficiently search the data. The server periodically updates this data to keep it up to date. The input is existing approval and guideline data, and the output is an indexed database.

[1874] Step 2:

[1875] The user opens the approval proposal form on the terminal and enters details such as the approval title, purpose, content, budget, schedule, etc. The terminal receives the user's input and sends the input data to the server. The input is the user's approval details, and the output is the data sent to the server.

[1876] Step 3:

[1877] The server receives the approval data entered by the user and searches the database for related past approval data and guideline data. The server sends the search results to the terminal and instructs it to display them. The input is the user's approval data and the database data, and the output is the search results for related past approval data and guideline data.

[1878] Step 4:

[1879] The terminal displays past approval data and guideline data received from the server for the user's reference. The user inputs and modifies the details of the approval while referring to the displayed data. The input is data from the server, and the output is displayed to the user.

[1880] Step 5:

[1881] After the user has finished entering the details of the approval request, they click the "Check" button on their device. The device then sends the user's approval request to the server. The input is the user's check request and the approval request details, and the output is the data sent to the server.

[1882] Step 6:

[1883] The server analyzes the received approval request and automatically checks it using a generative AI model. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The server generates correction suggestions based on the prompt text and sends the results to the terminal. The input is the user's approval request content and the generative AI model, and the output is correction suggestions and check results.

[1884] Step 7:

[1885] The terminal displays the correction suggestions and check results received from the server for the user to review. The user then corrects the approval request based on the feedback provided. The input is data from the server, and the output is the display and feedback to the user.

[1886] Step 8:

[1887] After the user has finished making the corrections, they click the "Check" button again, and the terminal sends the approval request content to the server again. The input is the approval request content after the user's corrections, and the output is the re-transmission to the server.

[1888] Step 9:

[1889] The server analyzes the approval request content again and checks whether all check items have been cleared. If the analysis results are satisfactory, it generates feedback stating "All items can be approved" and sends it to the terminal. The input is the revised approval request content and the generation AI model, and the output is the final feedback.

[1890] Step 10:

[1891] The user checks the final feedback and, if there are no problems, clicks the "Final Application" button. The terminal then sends the final approval request to the server. The input is the user's approval request and the final approval request, and the output is the final data sent to the server.

[1892] Step 11:

[1893] The server sends the received final approval content to the superior approver and starts the approval process. The input is the final approval content, and the output is sending the data to the superior approver and starting the approval process.

[1894] In this way, the process of drafting and reviewing approval requests is made more efficient, and quality is improved.

[1895] 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.

[1896] The embodiments of the present invention will be specifically described below.

[1897] The system of the present invention works in cooperation with the server, the terminal, and the user to improve the efficiency and quality of the approval process.The present invention also includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[1898] System configuration

[1899] 1. Server:

[1900] The server stores past approval data and guideline data, and manages the database so that they can be searched and referenced efficiently.

[1901] The server analyzes the content of the approval request sent by the user and automatically checks it based on guidelines and past approval request data.

[1902] Check results and correction suggestions are generated and sent to the user's device.

[1903] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[1904] 2. Terminal:

[1905] The terminal provides an input form for the user to draft a request for approval.

[1906] When a user submits a proposal, relevant past proposal data and guideline data are displayed.

[1907] It has a check execution button and a final application button, and has the function of sending the user's approval request details to the server.

[1908] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[1909] 3. User:

[1910] Users use their devices to draft requests and revise them based on feedback from the server.

[1911] The final revised approval request is sent to the higher approver.

[1912] The emotion engine receives input through the device's camera and sensors.

[1913] Program processing flow

[1914] 1. Prepare the database

[1915] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes it for efficient searches.

[1916] 2. Start of proposal process

[1917] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[1918] The terminal sends a request to the server based on the "category" field entered by the user at the time of drafting.

[1919] 3. Display of past approval requests and guidelines

[1920] The server receives the request and retrieves data related to the subject's classification from a database and returns it.

[1921] The terminal displays this data for the user to refer to.

[1922] 4. Run an automatic check

[1923] After the user has finished writing the details of the request, he / she clicks the "Check" button.

[1924] The server receives the approval request and automatically analyzes it to check its accuracy and completeness based on specific criteria and rules.

[1925] The server generates check results including points that need to be corrected and points to note, and sends these to the terminal.

[1926] 5. Feedback adjustment

[1927] The device's emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. For example, if the user is feeling stressed, the device will provide more polite and detailed feedback.

[1928] The server adjusts the content and tone of the feedback based on information from the emotion engine.

[1929] 6. Reviewing and Correcting Feedback

[1930] The user checks the feedback from the server on the device, confirms the points pointed out and points to note, and makes any necessary corrections.

[1931] After making the corrections, click the "Check" button again to request the server to check again.

[1932] 7. Final application execution

[1933] The server will recheck to make sure all checkpoints are cleared.

[1934] The user receives feedback from the server that "all items are acceptable" and clicks the final application button.

[1935] The server sends the final approval to the higher approver, and the approval process begins.

[1936] Specific examples

[1937] As an example, consider the case where a user submits a request for approval for a "new product development project."

[1938] 1. User Tanaka opens the approval proposal form on his / her device and selects "New Product Development Project."

[1939] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[1940] 3. Using these as a reference, Tanaka fills out the details of the request (purpose, required budget, schedule, etc.).

[1941] 4. After filling out the form, Tanaka clicks the "Check" button.

[1942] 5. The server analyzes the request and returns feedback saying, "There is insufficient budget item detail."

[1943] 6. The device's emotion engine detects that Tanaka is feeling a little stressed.

[1944] 7. The server then generates more detailed suggestions for correction and feedback in a gentler tone and sends it to Tanaka.

[1945] 8. Tanaka reviews the feedback and makes revisions, such as specifying budget items.

[1946] 9. Click the "Check" button again to request a recheck of the server.

[1947] 10. After all items have been cleared, Tanaka clicks the "Final Application" button to send the final approval request details to the server.

[1948] 11. The server sends the request to the higher approver and the approval process is completed.

[1949] In this way, the system of the present invention realizes improved efficiency and quality in the process of drafting approval requests, taking into account the emotional state of the user.

[1950] The processing flow will be explained below.

[1951] Step 1:

[1952] The server stores past approval and guideline data in a database, including information such as project name, purpose, budget, applicant, approval date, and classification. The data is indexed to enable efficient searches.

[1953] Step 2:

[1954] The user opens the request draft form on their device, selects the appropriate "category," and begins entering details such as the request title, purpose, content, budget, and schedule.

[1955] Step 3:

[1956] Based on the "category" selected by the user, the terminal sends a request to the server to search for related past approval requests and guideline data.

[1957] Step 4:

[1958] The server receives the request and retrieves data related to the target classification from a database and returns it.

[1959] Step 5:

[1960] The terminal displays the acquired past approval data and guideline data so that the user can refer to it, allowing the user to fill out the approval content while referring to it.

[1961] Step 6:

[1962] After the user has finished entering the details of the approval request, they click the "Check" button, which sends the details of the approval request to the server.

[1963] Step 7:

[1964] The server analyzes the received approval request and compares it with the guidelines and past approval request data to check the accuracy and completeness of the content based on specific standards and rules.

[1965] Step 8:

[1966] To recognize the user's emotional state, the device uses cameras and sensors to analyze the user's facial expressions and tone of voice using an emotion engine.

[1967] Step 9:

[1968] The server generates the analysis results and adjusts the feedback based on the information from the emotion engine, for example, if the user is feeling stressed, the feedback will be presented in a gentler tone.

[1969] Step 10:

[1970] The server generates the check results and correction suggestions and sends them to the terminal.

[1971] Step 11:

[1972] The user receives the feedback, checks the points and points of caution on the device, and makes any necessary corrections.

[1973] Step 12:

[1974] After making the corrections, the user clicks the "Check" button again to request a recheck from the server.

[1975] Step 13:

[1976] The server re-analyzes the received request and checks whether all check points have been met. If there are no problems, it generates feedback stating "All items are acceptable."

[1977] Step 14:

[1978] The user checks the final feedback and clicks the "Final Submission" button. The final approval request is sent to the server.

[1979] Step 15:

[1980] The server performs a final check and sends the request to the higher approver, after which the approval process begins.

[1981] Example 2

[1982] 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."

[1983] In conventional approval proposal systems, users had to take a lot of time and effort to submit approvals, and the quality of the approvals was inconsistent. This could also cause stress and strain on users, making it difficult to create efficient, high-quality approval proposals. Furthermore, because feedback was uniform, flexible support that took into account the user's emotional state was not possible.

[1984] 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.

[1985] In this invention, the server includes a means for storing and retrieving past approval data and guideline data, a means for automatically checking the content of a user's approval and suggesting corrections and points to check, and a means for recognizing the user's emotional state and adjusting the method and content of feedback. This improves the efficiency of the user's approval drafting process and the quality of approval content. Furthermore, it is possible to reduce the user's stress and burden and provide flexible support according to the user's emotional state.

[1986] "Past approval data" refers to information related to approval documents that have previously been created or approved, which is stored in a database and is searchable.

[1987] "Guideline data" refers to documents and data that describe standard or recommended procedures, conditions, and standards for drafting approvals.

[1988] "Classification" refers to categories for organizing approval data and guideline data by purpose, project, or other criteria.

[1989] "Automatic checking" refers to the process in which the system analyzes the contents of a user's request and verifies it against pre-defined standards and rules.

[1990] "Points to correct and check" indicate the parts of the user's approval request that need to be corrected or items that should be checked, as identified as a result of the automated check.

[1991] "Emotional state" refers to the user's current psychological or emotional state, and is estimated from facial expressions, tone of voice, and the like.

[1992] "Feedback" refers to information and advice provided to the user by the system, which is useful for correcting and improving the content of the request.

[1993] "Superior approver" refers to the person who has the authority to make the final evaluation of the contents of the request and decide whether to approve or reject it.

[1994] "Indexing" refers to the process of assigning search keys to specific items so that information in a database can be efficiently searched.

[1995] A specific embodiment for carrying out the present invention will now be described. The system of the present invention works in cooperation with a server, a terminal, and a user to improve the efficiency and quality of drafting approval requests. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it also includes a function that provides flexible support.

[1996] System configuration

[1997] 1. Server:

[1998] The server manages a searchable database that stores past approval and guideline data, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data to enable efficient searches.

[1999] The server analyzes the content of the approval request submitted by the user and automatically checks it based on guidelines and past approval request data. Specifically, it uses a generative AI model to check the accuracy and completeness of the content.

[2000] The server generates the check results and correction suggestions and sends them to the user's device. This feedback is provided in a way that takes into account the user's emotional state. The content and tone of the feedback are adjusted based on information from the emotion engine.

[2001] Send the final approval request to a higher approver to begin the approval process.

[2002] 2. Terminal:

[2003] The terminal provides an input form for the user to submit a request for approval, which includes fields for entering details such as the title, purpose, content, budget, and schedule of the request.

[2004] When a user enters data, the system has the function of requesting, retrieving, and displaying past approval data and guideline data so that the user can refer to this data.

[2005] It has a function to set up a check execution button and a final application button, and to send data to the server according to the user's operation.

[2006] The emotion engine uses sensors such as cameras and microphones to analyze the user's facial expressions and tone of voice to estimate their emotional state. This information is sent to a server and used to adjust feedback.

[2007] 3. User:

[2008] Users use their devices to draft requests and revise them based on feedback sent back from the server.

[2009] Finally, the revised approval request is sent to the higher approver.

[2010] The device's camera and sensors provide input to the emotion engine, for example, entering approval details, running checks, and reviewing and correcting feedback.

[2011] Specific examples

[2012] For example, consider the case where a user submits a request for approval for a "new product development project."

[2013] 1. The user opens the approval proposal form on their device and selects "New Product Development Project."

[2014] 2. The terminal sends this information to the server, and retrieves and displays past approval requests and guideline data for the same category.

[2015] 3. The user uses the acquired data to fill out the details of the request (purpose, required budget, schedule, etc.).

[2016] 4. After filling out the form, the user clicks the "Check" button.

[2017] 5. The server analyzes the request and generates feedback saying "There is insufficient budget item details."

[2018] 6. The device's emotion engine analyzes the user's facial expressions and tone of voice to detect that the user is feeling slightly stressed.

[2019] 7. Based on this emotional information, the server generates more detailed suggestions for correction and polite feedback and sends it to the user.

[2020] 8. The user reviews the feedback and makes corrections, such as specifying budget items.

[2021] 9. Click the "Check" button again to request the server to check again.

[2022] 10. After all items have been cleared, the user clicks the "Final Application" button to send the final approval request to the server.

[2023] 11. The server sends the request to the higher approver and the approval process begins.

[2024] In this way, the system takes into account the user's emotional state and realizes efficient and high-quality drafting of approval requests.

[2025] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2026] Step 1: Prepare the Database

[2027] The server stores past approval and guideline data in a database, including approval classification, project name, purpose, budget, applicant, approval date, etc., and indexes the data. The input is the previously collected approval and guideline data, and the output is a database that can be efficiently searched.

[2028] Step 2: Initiate the drafting process

[2029] The user opens the approval proposal form on the terminal, selects the project category, and enters the title, purpose, content, budget, schedule, etc. of the approval. Using the user's selected category and detailed approval content information as input, the terminal sends this information to the server. As output, the request for approval content sent to the server is obtained.

[2030] Step 3: View past approvals and guidelines

[2031] Based on the received request, the server searches and retrieves data related to the target classification from the database. As input, it uses the user's submitted data and search conditions in the database, and as output, it obtains data on past approval requests and guideline data. It then sends this data to the terminal. The terminal displays this data so that the user can refer to it.

[2032] Step 4: Run automated checks

[2033] After the user has finished filling out the request, they click the "Check" button. The completed request is used as input, and the server receives this data and automatically analyzes it using a generative AI model. The output is a check of the accuracy and completeness of the request, resulting in feedback including corrections and points to check, which is sent to the device.

[2034] Step 5: Adjusting feedback

[2035] The device's emotion engine uses a camera and microphone to analyze the user's facial expressions and tone of voice to estimate their emotional state. The user's facial expression and voice data are used as input, and the estimated emotional state is obtained as output. The server then adjusts the tone and content of the feedback based on this emotional state.

[2036] Step 6: Review and correct feedback

[2037] The user checks the feedback from the server on the terminal. The displayed feedback is used as input, and the user corrects the pointed out parts. After checking the corrections, the user clicks the "Check" button again to request a recheck from the server. As output, a request with the corrected approval content is sent to the server.

[2038] Step 7: Submit the final application

[2039] The server rechecks and confirms that all check points have been cleared. The revised approval request content is used as input, and feedback that it can be approved is sent to the terminal as output. The user clicks the "Final Application" button to send the final approval request content to the server. The server sends this final approval request content to the superior approver, and the approval process begins. The output is a status that the application to the superior approver has been completed.

[2040] (Application example 2)

[2041] 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."

[2042] Conventional approval proposal systems are inefficient in checking and correcting approval content, and do not provide feedback that takes into account the user's emotional state, which can lead to stress and reduced efficiency. Furthermore, similar problems can occur in the management of work instructions in factories and on-site, potentially resulting in reduced work efficiency and content quality.

[2043] 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.

[2044] In this invention, the server includes means for storing past approval request data and guideline data and making them efficiently searchable and accessible, means for displaying the past approval request data and guideline data so that they can be referenced on an input device used by a user to draft an approval request, means for automatically checking the content of the user's approval request and presenting corrections and points to be checked, means for rechecking the approval request content revised by the user based on the corrections and points to be checked, means for sending the final approval request content to the approver, and means for recognizing the user's emotional state and adjusting the content and tone of the feedback. This enables flexible feedback that takes the user's emotional state into consideration, improving the efficiency and quality of approval request drafting and work instruction management.

[2045] The "server" is a device that stores past approval data and guideline data and makes them available for efficient search and reference.

[2046] The "database" is a system for efficiently managing, storing, and accessing past approval data and guideline data.

[2047] A "terminal" is an input device that allows a user to draft a request for approval and input and edit its contents.

[2048] The "emotion engine" is a system that analyzes a user's facial expressions and tone of voice to estimate their emotional state.

[2049] A "rei-i" is a document that records and processes proposals, applications, and approval processes within an organization.

[2050] The "check execution button" is an input means by which the user requests the server to automatically check the contents of the approval request.

[2051] The "final application button" is an input means for the user to send the final approval request content to the approver.

[2052] An "approver" is a person or system that has the authority to review the contents of the request and make final approval.

[2053] "Points to be corrected" are areas that require correction and are discovered when checking the contents of the request.

[2054] "Verification points" are areas that require verification and are discovered when checking the contents of the approval request.

[2055] "Classification" refers to the categories used to organize and index approval data and guideline data.

[2056] An "index" is an index that enables efficient reference to data.

[2057] To implement this invention, it is necessary to build a system that makes work instruction management more efficient and improves its quality through collaboration between the server, the terminal, and the user. This system includes a function that provides more flexible support according to the user's situation by combining it with an emotion engine that recognizes the user's emotions.

[2058] System configuration

[2059] 1. Server:

[2060] The server stores past work instruction data and guideline data and manages the database so that they can be efficiently searched and referenced.

[2061] The server analyzes the contents of the work instructions sent by the user and automatically checks them based on guidelines and past data.

[2062] Check results and correction suggestions are generated and sent to the user's device.

[2063] An emotion engine is used to recognize the user's emotional state and adjust the method and content of feedback.

[2064] 2. Terminal:

[2065] The terminal provides an input form for the user to create a work instruction.

[2066] Users can view relevant historical and guideline data when creating work instructions.

[2067] It has a check execution button and a final send button, and has the function of sending the contents of the user's work instructions to the server.

[2068] It is equipped with an emotion engine and uses cameras and sensors to analyze the user's facial expressions and tone of voice to estimate their emotional state.

[2069] 3. User:

[2070] The user uses the terminal to create work instructions and revise the contents based on feedback from the server.

[2071] The final revised work instruction is sent to a higher approver.

[2072] The emotion engine receives input through the device's camera and sensors.

[2073] Program processing flow

[2074] The server stores past approval data and guideline data in a database, allowing users to efficiently refer to related data when drafting work instructions on their devices. Users enter the details of the work instructions through an input form, and the server performs automatic checks based on this. During this process, an emotion engine identifies the user's emotional state and generates appropriate feedback, which is sent to the device. This allows users to efficiently create and modify work instructions while receiving flexible support according to their emotional state.

[2075] Equipment and software used

[2076] Hardware: Camera (e.g. USB camera), device (e.g. tablet or PC)

[2077] Software: OpenCV (face detection), TensorFlow (emotion recognition), custom feedback system

[2078] Specific examples

[2079] Consider the case of a factory line leader creating work instructions for a new production line. Using this system, the following steps are streamlined:

[2080] 1. The user creates a work instruction on the terminal.

[2081] 2. Send the work order to the server and perform an automatic check.

[2082] 3. The emotion engine analyzes the user's facial expressions and tone of voice and provides feedback according to their emotional state.

[2083] 4. The user modifies the work order based on the feedback and submits it to the server again.

[2084] 5. Click the final submit button to send the work order to the approver.

[2085] Prompt Sentence Examples

[2086] Enter the following information into your factory's work order creation support system:

[2087] 1. Work title

[2088] 2. Details of the work

[2089] 3. Materials needed

[2090] 4. Work Procedure

[2091] 5. Safety Guidelines

[2092] 6. User's real-time emotional state

[2093] This allows for flexible feedback that takes into account the user's emotional state, improving the efficiency and quality of work instruction management.

[2094] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2095] Step 1:

[2096] The server stores historical work instruction data and guideline data in a database, including work classification, project name, purpose, required materials, procedures, creators, etc., and indexes it to enable efficient searches. The input is historical work instruction data and guideline data, and the output is the indexed database.

[2097] Step 2:

[2098] The user opens the work instruction draft form on the terminal, selects the appropriate "classification," and begins to enter details such as the work title, purpose, content, required materials, procedures, schedule, etc. The input is the user's work instruction information, and the output is the draft work instruction that the user entered and saved on the terminal.

[2099] Step 3:

[2100] The device sends a request to the server based on the "Classification" field entered by the user. The server receives the request, retrieves data related to the target classification from the database, and returns it to the device. The input is the classification request made by the user, and the output is related historical data and guidelines.

[2101] Step 4:

[2102] After the user has finished writing the work instructions, he / she clicks the "Check" button. The terminal sends the work instructions to the server, which receives them. The input is the work instructions written by the user, and the output is the work instructions data received by the server.

[2103] Step 5:

[2104] The server analyzes the received work instruction content and automatically checks it. The analysis checks the accuracy and completeness of the content based on specific criteria and rules. The input is the work instruction content, and the output is the check results and correction suggestions based on specific criteria.

[2105] Step 6:

[2106] The emotion engine analyzes the user's facial expressions and tone of voice to infer their emotional state. The device does this using cameras and sensors. The input is the user's real-time facial and voice data, and the output is the inferred emotional state.

[2107] Step 7:

[2108] The server adjusts the content and tone of the feedback based on information from the emotion engine. Depending on the emotional state, it refines the necessary correction suggestions and generates feedback that is considerate to the user. The input is the result of emotion recognition, and the output is the adjusted feedback content.

[2109] Step 8:

[2110] The user checks the feedback from the server on the terminal, confirms the points pointed out and points to note, and makes the necessary corrections. The input is the server's feedback, and the output is the corrected work instructions.

[2111] Step 9:

[2112] The user clicks the "Check" button again to request a recheck from the server. The server performs the recheck and confirms that all check points have been cleared. The input is the work instruction modified by the user and the recheck request, and the output is the result of the recheck.

[2113] Step 10:

[2114] After all items are cleared, the user clicks the "Final Submission" button and sends the final work order content to the server. The server sends the final work order content to the approver, and the approval process is then carried out. The input is the final modified work order, and the output is the work order sent to the approver.

[2115] 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.

[2116] 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.

[2117] 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.

[2118] 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.

[2119] 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.

[2120] 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.

[2121] 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).

[2122] 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 ...

Claims

1. A means of storing and making searchable past approval and guideline data; a display means for displaying the past approval request data and guideline data so that the user can refer to them on the terminal where the approval request is submitted; A means to automatically check the contents of the user's approval request and suggest corrections and points to check, A means for rechecking the content of the approval request corrected by the user based on the corrections and confirmation points; A system that includes a means to send the final approval content to a higher approver.

2. The system according to claim 1, further comprising means for organizing and indexing past approval data and guideline data by the same classification.

3. The system according to claim 1, further comprising means for presenting detailed suggestions for correction and points to note based on the results of the automatic check when a user drafts a request for approval.

Images