Information processing device, information processing method, information processing program, and information processing system

Abstract

The system reduces the burden on users performing a given task by appropriately extracting the information they need from content that includes work information related to that task, and by reducing the burden of requesting items such as parts and tools required for that task. [Solution] The information processing device performs the following processes: identifying a specific task from among multiple tasks associated with each of the multiple task information contained in the first content; extracting item information of items related to the tasks in the task information contained in the first content; and generating a second content which includes at least a portion of the task information related to the identified task and includes information that makes it possible to request items corresponding to the extracted item information.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, an information processing method, an information processing program, and an information processing system.

Background Art

[0002] In recent years, a technique for summarizing predetermined document content using a learned model created by machine learning is known. For example, in Patent Document 1, as a method for creating a summary of a research paper, the document content is decomposed into sentences, labels corresponding to the content of the sentences are assigned using a learned model, and sentences assigned labels corresponding to the purpose or significance of the research are extracted as summary elements, and a summary based on the summary elements is created using a learned model.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is content (e.g., "service manual (work guide, work procedure manual)") that defines the implementation in a predetermined operation or operation sequence in order to provide a service (e.g., vehicle maintenance service, etc.) with a certain quality. Such a service manual is for providing a service with a certain quality as described above, and although it is recommended that it be described in an easy-to-understand and detailed manner for the service provider (hereinafter referred to as the user), the appropriate service response content may vary depending on various factors such as the service-providing environment, facilities, and time zone. Also, depending on the type of service provided, there are those (special operations) whose operation content and procedures are complicated, and service manuals regarding these contain a huge amount of information (types of information, pages).

[0005] Proper use of service manuals is necessary to appropriately provide services that maintain a certain level of quality. However, as mentioned above, if service manuals are complex and voluminous, it is difficult for users to understand their contents in advance. Furthermore, even if users read through the service manual on the spot, it takes time. On the other hand, when creating a summary of a service manual using the method disclosed in Patent Document 1, it is not possible to create a summary of the service manual that is limited to, for example, the parts related to the services provided by the user.

[0006] Furthermore, when performing work according to the service manual, parts, tools, and other items may be required. Traditionally, users had to check the service manual to determine what items were needed, and then place separate orders or make requests via telephone, fax, etc., which was a burden on the user.

[0007] This disclosure is made in view of the above-mentioned issues. Specifically, the purpose of this disclosure is to reduce the burden on users performing a given task by appropriately extracting the information they need from content, including work information related to that task, and by reducing the burden of requesting items such as parts and tools necessary for that task. [Means for solving the problem]

[0008] An information processing apparatus according to one aspect of the present disclosure is an information processing apparatus comprising at least one processor and a storage device for storing an information processing program, wherein the processor executes the information processing program to perform: a user input acquisition process for acquiring user input information from a user terminal used by a user; a user information acquisition process for acquiring user information based on the user input information by referring to a user database for storing user information relating to the user; a task identification process for identifying a specific task from among the multiple tasks related to the multiple task information included in the first content, based on the user information or the user input information, by referring to a first content database for storing a plurality of task information each related to a plurality of tasks; an item extraction process for extracting item information of tasks related to the task information included in the first content; a second content generation process for generating a second content which includes at least a part of the task information relating to the identified task and information that enables the request of an item corresponding to the extracted item information, based on the task information relating to the specific task and the extracted item information; and an output process for outputting the second content to the user terminal.

[0009] An information processing method according to one aspect of the present disclosure is an information processing method performed by an information processing device comprising at least one processor and a storage device for storing an information processing program, wherein the information processing device includes: a user input acquisition step of acquiring user input information from a user terminal used by a user; a user information acquisition step of acquiring user information based on the user input information by referring to a user database that stores user information relating to the user; a task identification step of identifying a specific task from among the multiple tasks related to the multiple task information included in the first content based on the user information or the user input information by referring to a first content database that stores a plurality of task information each related to a plurality of tasks; an item extraction step of extracting item information of tasks related to the task information included in the first content; a second content generation step of generating a second content which includes at least a part of the task information relating to the identified task and information that makes it possible to request an item corresponding to the extracted item information, based on the task information relating to the specific task and the extracted item information; and an output step of outputting the second content to the user terminal.

[0010] An information processing program according to one aspect of the present disclosure is an information processing device that is executed in an information processing device comprising at least one processor and a storage device for storing the information processing program, wherein when the information processing program is executed by the processor, the information processing device is caused to execute: a user input acquisition process for acquiring user input information from a user terminal used by a user; a user information acquisition process for acquiring user information based on the user input information by referring to a user database for storing user information relating to the user; a task identification process for identifying a specific task from among the multiple tasks related to the multiple task information included in the first content, based on the user information or the user input information, by referring to a first content database for storing a first content containing a plurality of task information each related to a plurality of tasks; an item extraction process for extracting item information of tasks related to the task information included in the first content; a second content generation process for generating a second content that includes at least a part of the task information relating to the identified task and information that enables the request of an item corresponding to the extracted item information, based on the task information relating to the specific task and the extracted item information; and an output process for outputting the second content to the user terminal.

[0011] An information processing system according to one aspect of the present disclosure performs the following: a user input acquisition process that acquires user input information from a user terminal used by a user; a user information acquisition process that acquires user information based on the user input information by referring to a user database that stores user information about the user; a task identification process that identifies a specific task from among the multiple tasks related to the multiple task information included in the first content based on the user information or the user input information by referring to a first content database that stores a plurality of task information each related to a plurality of tasks; an item extraction process that extracts item information of items related to tasks in the task information included in the first content; a second content generation process that generates a second content which includes at least a part of the task information related to the identified task and information that makes it possible to request an item corresponding to the extracted item information, based on the task information related to the specific task and the extracted item information; and an output process that outputs the second content to the user terminal. [Effects of the Invention]

[0012] According to this disclosure, the burden on the user of a given task can be reduced by appropriately extracting the information the user needs from content including work information related to that task, and by reducing the burden of requesting items such as parts and tools necessary for that task. [Brief explanation of the drawing]

[0013] [Figure 1] Figure 1 is a schematic diagram showing an example of a system according to the present disclosure. [Figure 2] Figure 2 is a block diagram showing an example of the functional configuration of the system according to the embodiment of this disclosure. [Figure 3] Figure 3 is an example of the information stored in the user database. [Figure 4] Figure 4 is an example of the information stored in the correspondence table. [Figure 5] FIG. 5 is a diagram illustrating a management table for managing a plurality of first contents stored in work information. [Figure 6] FIG. 6 is a diagram illustrating the first content. [Figure 7] FIG. 7 is a flowchart schematically showing an example of an information processing method according to an embodiment of the present disclosure. [Figure 8] FIG. 8 is a flowchart showing an example of the second content related processing shown in FIG. 7. [Figure 9] FIG. 9 is a flowchart showing an example of the extraction processing of important information shown in FIG. 8. [Figure 10] FIG. 10 is a diagram illustrating the second content displayed on the user terminal. [Figure 11] FIG. 11 is a diagram illustrating the second content generated by a generation method different from the example of FIG. 10. [Figure 12] FIG. 12 is a flowchart showing an example of the request support processing shown in FIG. 7. [Figure 13] FIG. 13 is a diagram illustrating the information stored in the item management DB and the related parts DB. [Figure 14] FIG. 14 is a schematic diagram showing an example of an order screen. [Figure 15] FIG. 15 is a schematic diagram showing an example of an order screen. [Figure 16] FIG. 16 is a schematic diagram showing an example of an order screen. [Figure 17] FIG.​​​​​​​​​​​​​​Figure 22 is an example of the information stored in the user database. [Figure 23] Figure 23 is a schematic diagram showing an example of a checklist. [Figure 24] Figure 24 is a schematic diagram showing an example of the third content. [Modes for carrying out the invention]

[0014] Hereinafter, examples of embodiments of this disclosure will be described with reference to the drawings. The same reference numerals are used for identical components, and redundant explanations will be omitted as appropriate. Furthermore, configurations and processes not related to the characteristic features of the embodiments may be omitted.

[0015] In the following, as one embodiment of this disclosure, we will describe a system that includes an information processing device capable of extracting important information from a first content containing multiple work information related to multiple tasks, generating a second content based on the extracted important information, and outputting the second content to a user terminal operated by the user. Furthermore, the system described below is capable of providing user request support (support for various request actions such as placing orders) and evidence support, and these will also be described in this embodiment.

[0016] First, let's explain each term used in this specification. Below, we will mainly explain the case where "work" refers to work related to the maintenance and repair of vehicles (automobiles), but "work" is not limited to this example and can refer to any type of work. Examples of "work" include maintenance, repair, operation, setup, and production of objects (for example, transportation equipment such as automobiles, motorcycles, trains, ships, and airplanes; electronic devices such as personal computers and smartphones; home appliances such as televisions and refrigerators; machine tools such as lathes; and various parts related to these devices).

[0017] A "user" is anyone who performs the tasks described above. Examples of "users" include maintenance companies, repair shops, manufacturers, and setup companies that perform the above tasks as a business. Alternatively, a "user" may be an ordinary person who performs the above tasks on a personal basis.

[0018] "Primary content" refers to content that includes multiple pieces of work information related to multiple tasks (for example, work procedures and methods, necessary parts, points to note, etc.). Examples of "primary content" include service manuals (work guides, work procedure manuals, etc.), instructions, recall information, parts information, and instruction manuals provided by the manufacturer of the object, as well as reference information such as know-how held by the operator of System 5, and various information provided by other external organizations (for example, government agencies or industry associations). "Primary content" includes, and may include, text, illustrations, photographs, tables, videos, audio, etc.

[0019] "Second Content" includes, for example, at least a portion of the work information included in the first content that pertains to a specific task (for example, a task that the user should perform). Preferably, "Second Content" is content that includes, for example, at least a portion of the work information pertaining to a specific task, and includes information that enables the presentation of a request screen that can request items that can be used for that task. Preferably, "Second Content" is content related to the first content, created based on important information extracted from the work information included in the first content. Examples of "Second Content" include a summary type that summarizes the first content, a point-by-point type that lists important information in a predetermined order, an information-added type that adds important information or supplementary information related to important information to the first content or its summary, and a difference type that shows only the changes made to the first content or highlights the changes made when the first content has been modified. "Second Content" includes, of course, text, illustrations, photographs, tables, videos, audio, etc., and may include multiple of these.

[0020] An "item" is a general term for supplies and various objects that can be used in a task. Items can be classified according to their nature and purpose into, for example, parts to be replaced or installed, tools used in the work, and other supplies. In the field of automotive maintenance, specific examples of items to be replaced or installed include various parts (engine parts, brake pads, filters, belts, electrical components, etc.), various consumables (oil, LLC (coolant), brake fluid, grease, etc.), and various materials (paints, adhesives, sealants, welding materials, etc.). Tools used in the work include, for example, general-purpose hand tools (wrenches, screwdrivers, pliers, hammers, etc.), general-purpose power tools (impact wrenches, grinders, drills, etc.), specialized automotive tools (timing belt replacement tools, suspension tools, etc.), measuring instruments (torque wrenches, micrometers, testers, diagnostic tools (scan tools), etc.), and equipment and machinery (lifts, jacks, presses, welding machines, paint booths, etc.). Other supplies include, for example, protective equipment (gloves, safety glasses, work clothes, etc.) and cleaning supplies (rags, cleaning agents, parts cleaner, etc.). These various items are sometimes referred to as maintenance supplies in the field of automotive maintenance. However, the term "items" is not limited to items in the field of automotive maintenance.

[0021] Furthermore, the term "request" for items, as described later, includes not only ordering items but also requests for quotations, inventory inquiries, special orders, reservations, and various other requests related to obtaining items. Therefore, "requesting an item" includes ordering an item, requesting a quotation for an item, inquiring about the item's inventory, requesting an item to be specially ordered, and reserving an item. In this context, "recipient of the request" includes the supplier, the recipient of the quotation, the recipient of the inventory inquiry, the recipient of the special order request, and the recipient of the reservation.

[0022] "Ordering" refers to a series of information processing steps in which the information processing device 1 generates order request data to request the supply of an item based on the operation input received from the user terminal 2, and transmits the order request data to the requesting device (ordering device) 7 via the network 4. Completion of the order processing means that the transmission of the order request data from the information processing device 1 to the requesting device 7 is completed successfully, and response data indicating order acceptance is received from the requesting device 7. Once the order processing is complete, for example, the requesting device 7 starts processing to ship the ordered item.

[0023] The order request data may include at least one of the following: order target information, which includes information to identify the item to be ordered (item ID, model number, name, etc.); order condition information, which includes information indicating conditions related to the order, such as quantity information, ordering source information, desired delivery date, payment terms, and delivery method; and may further include related work information, which is information to identify the work that triggered the order (project ID, work content ID, etc.).

[0024] "Related items" refer to items that are interchangeable with or can be used simultaneously with an item. Related items are defined for each item, and there can be multiple related items for a single item. Also, one item may be a related item to another item. To give a specific example in the field of automotive maintenance, for example, genuine parts (parts officially sold by the vehicle manufacturer), second-hand genuine parts (parts officially sold by parts retailers), high-quality parts (parts that are identical to genuine parts but are sold by parts manufacturers, not by the vehicle manufacturer, and only differ from genuine parts in model number), used parts (parts removed from scrapped vehicles, etc., that have been repaired to make them sellable), and compatible parts (parts that fit multiple vehicle models, regardless of whether they are new or used) can all be related items to each other. Furthermore, if the item is a part, etc. that is to be replaced or installed, for example, the tools used in the work may be related items. In the following explanation, items and related items will be collectively referred to as "items, etc."

[0025] Furthermore, the "history information" described later refers to record information about the work performed by the user after referring to the second content. This "history information" includes, for example, the date and time of the work (start date and / or end date and time of the work), the work location (business office where the work was performed, pit number, etc.), information about the equipment used for the work (identification information for lifts, diagnostic equipment, tools, etc. used), user identification information to identify the user who performed the work (user ID, name, etc.), and the specific details of the work (inspection items, replacement parts, adjustment values, etc.). The "history information" may also include information about the items used in the work, link information to videos or still images of the work situation, identification information of the vehicle that was the target of the work (chassis number, etc.), and the work results (pass / fail, measured values, etc.).

[0026] Furthermore, the "task completion information" described later refers to information indicating that a user has completed a predetermined task. "Task completion information" is generated based, for example, on a completion report entered by the user operating the user terminal, a completion determination result automatically determined from video or still images of the user's work status, or the result of detecting a change in the state of the object of the work (vehicle, etc.).

[0027] Figure 1 is a schematic diagram showing an example of a system 5 according to the embodiment of this disclosure. As shown in Figure 1, the system 5 includes an information processing device 1 and a plurality of user terminals 2. The information processing device 1 and the user terminals 2 can communicate with each other via a network 4 and can send and receive data from each other.

[0028] The information processing device 1 is a computer that performs various information processing. In this embodiment, the information processing device 1 functions as a server that outputs second content to the user terminal 2 in response to a request from the user terminal 2. The information processing device 1 may be a cloud-type server or an on-premise server. The information processing device 1 includes, for example, a processor 11, memory 12, storage 13, and communication I / F 14. The processor 11, memory 12, storage 13, and communication I / F 14 are connected to each other, for example, via a communication bus. The information processing device 1 may also include other conventionally known components.

[0029] The processor 11 performs various calculations and controls. The processor 11 may be a general-purpose processor such as a CPU (Central Processing Unit), or a dedicated processor such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). The memory 12 includes a storage device such as a ROM (Read Only Memory) that stores various programs, and a RAM (Random Access Memory) that provides a work area when various programs are executed.

[0030] The storage 13 includes, for example, a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The storage 13 only needs to have a storage area that is accessible to the information processing device 1, and may be configured to have a dedicated storage area outside the information processing device 1, for example. The information processing device 1, for example, uses the data read into the memory 12 to perform various information processing using the processor 11, and stores the obtained processing results in the storage 13 as needed. The communication I / F 14 includes various processing circuits and connection terminals for communicating with external devices via a communication network, and is configured to conform to various communication standards.

[0031] User terminal 2 is a terminal operated by the user. In this embodiment, user terminal 2 includes, but is not limited to, user terminal 2A, which is a stationary information terminal such as a desktop computer; user terminal 2B, which is a portable information terminal such as a smartphone or tablet; and user terminal 2C, which is a camera that captures images of the user's work, etc. User terminal 2 may be, for example, a notebook computer or a PDA (Personal Digital Assistant), or various cameras such as smart glasses, VR goggles, AR glasses, AR contacts, wearable cameras such as body cameras or head cameras, or stationary monitoring cameras. Furthermore, user terminal 2 may include only one type of the above-mentioned information terminals, or it may include multiple types.

[0032] External device 3 and requesting device 7 are computers owned by parties other than the operator of system 5. Information processing device 1, external device 3, and requesting device 7 can communicate with each other via network 4 and send and receive data from each other. By accessing external device 3, information processing device 1 can obtain supplementary information related to important information. In addition, by accessing requesting device 7, information processing device 1 can obtain inventory information and place orders for items sold and managed by requesting device 7. External device 3 and requesting device 7 may be devices that the operator of system 5 has designated in advance as access targets. In the example in Figure 1, there is only one external device 3 and one requesting device 7, but there may be multiple external devices 3 and requesting devices 7.

[0033] Figure 2 is a block diagram showing an example of the functional configuration of system 5 according to the embodiment of this disclosure. As shown in Figure 2, the information processing device 1 comprises a storage unit 20 and a control unit 30. The storage unit 20 is composed of, for example, a storage device such as the ROM, HDD, or SSD described above. The storage unit 20 stores, for example, a program 21, a user DB 22, a first content DB 23, a first trained model 26, a fluctuation DB 27, a second trained model 28, an item management DB (item management database) 29A, and a related parts DB (related parts database) 29B.

[0034] Program 21 is, for example, an information processing program for implementing various functions described later in the information processing device 1. In this embodiment, Program 21 is stored in a storage device such as ROM, which is a non-temporary computer-readable medium. Program 21 may also be stored in any non-temporary computer-readable medium such as an optical disc (CD, DVD, Blu-ray disc, etc.) or semiconductor memory (EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, etc.) and provided to the information processing device 1.

[0035] User DB22 is a database that stores user information about users. User DB22 stores user attributes, etc., associated with information that can uniquely identify a user, such as a user ID. The user information contained in User DB22 is updated based on, for example, user input information and user work history.

[0036] Figure 3 illustrates the information stored in the user database 22. In the example in Figure 3, each user ID stored in the user ID field 221 is associated with the following attributes of each user: skill level, affiliation ID, work environment, worker history, target vehicle, chassis number, and vehicle history.

[0037] The proficiency level field 222 stores the user's proficiency level for tasks that the user can perform. In the example in Figure 3, it is shown in three levels from "1" to "3," but there are no particular restrictions on how the proficiency level is represented. The proficiency level may be indicated, for example, by the number of years the user has been engaged in the work, or by the presence or absence of a prescribed qualification or the results of a prescribed test. The proficiency level may also be determined by the information processing device 1 based on the worker's history or other user information, or it may be set by the user themselves. Furthermore, the proficiency level may be set for each attribute of the work to be performed (e.g., inspection items, repair locations, types of parts to be replaced, difficulty of the work, etc.) or for each attribute of the object being worked on (e.g., manufacturer name, model, year of manufacture, etc.), or for each combination of work attributes and object attributes.

[0038] The Affiliation ID stored in the Affiliation ID field 223 is information that uniquely identifies the organization (company, business office, etc.) to which the user belongs. In the example in Figure 3, "001A" and "001B" indicate that the company is the same but the business office is different. "001A" and "002A" indicate that the company itself is different. "999" indicates that the user is an individual who does not belong to an organization that performs repairs or similar services as a business.

[0039] The work environment field 224 stores information about the work environment, such as equipment and tools available within the organization to which the user belongs, as well as equipment and tools owned by the user personally. In the example in Figure 3, user "001-1" and user "001-3" work for the same company but at different locations, resulting in different work environments.

[0040] The worker history field 225 stores information indicating the history of work performed by the user. The worker history may be, for example, a history of work performed within a predetermined period, or it may include a history of all work performed since user registration. The worker history may also include information on the number of times each work has been performed and the time elapsed since the last work. Furthermore, the worker history may be classified and stored according to the attributes of the work, the attributes of the object, and / or a combination thereof. The worker history field 225 is updated, for example, each time the user completes a task.

[0041] The target vehicle column 226 and the chassis number column 227 store information indicating the target vehicle, which is the object the user plans to work on, and the chassis number of that target vehicle, respectively. In addition, the vehicle history column 228 stores the vehicle history, which is a record of the work performed on the above target vehicle.

[0042] In the example in Figure 3, vehicles "AA001" and "AA002" are from the same manufacturer but are of different models. Furthermore, the chassis number is a unique identification number assigned to each vehicle. Even for the same vehicle model, the chassis numbers will be different.

[0043] The information stored in the target vehicle field 226, the chassis number field 227, and the vehicle history field 228 can be obtained, for example, based on user input information. Specifically, the chassis number and information about the owner of the vehicle to be worked on are received as user input information, and information associated with the received user input information can be obtained by referring to other databases that store information about the vehicle and information about the owner. The above other databases may be stored in the storage unit 20 or in the external device 3.

[0044] The user database 22 only needs to have at least one of the above-mentioned pieces of information associated with the user ID. In addition to the above-mentioned pieces of information, the user database 22 may also contain other information, such as the user's name and address, and login passwords. The user database 22 also stores historical information about the work performed by the user. This historical information is stored as a record for each task, and each record includes at least the date and time of the task, the location of the task, information about the equipment used for the task, the user ID of the user who performed the task, and the specific details of the task. Examples of equipment information include lifts, diagnostic equipment, tools, and measuring instruments. Including this equipment information in the historical information can be used for quality assurance of work and management of equipment operation.

[0045] Returning to the explanation of Figure 2, the user DB 22 may contain quotation information 22A and project information 22B. Quotation information 22A includes, for example, information indicating the quotation items (work details, required items, etc.). Quotation information 22A is created or acquired by the information processing device 1 based on user input and stored in the user DB 22. Project information 22B will be described in detail in a later paragraph using Figure 22.

[0046] The first content DB23 includes, for example, a correspondence table 24 and work information 25. The correspondence table 24 is a table for identifying a specific work from among multiple work associated with multiple work information contained in the first content. In the correspondence table 24, for example, the user's status, determined based on user information and / or user input information, is associated with a specific work.

[0047] Figure 4 illustrates the information stored in the correspondence table 24. The User Request field 241 stores the type of work the user wishes to perform. The User Request Details field 242 stores the details of the work the user wishes to perform. The Vehicle Name field 243 stores the vehicle name. The Required Work field 244 stores information about work that the user must perform. The Recommended Work field 245 stores information about work that is not required but is recommended to be performed. The Keyword field 246 stores keywords related to the required and recommended work. The keywords stored in the Keyword field 246 are used, for example, in the important information extraction process described later.

[0048] For example, the user operates a user terminal to select the task they intend to perform from a predetermined list of items such as "periodic inspection," "maintenance / repair," and "symptom diagnosis." Afterward, the user selects details of the task, such as "12-month inspection," "2-year vehicle inspection," or "brake pad replacement." The user also selects the vehicle model to which the task applies. This user input determines the user's status, which in turn determines the relevant row in the correspondence table 24, thereby identifying a specific task.

[0049] A specific task may consist only of tasks stored in the mandatory tasks column 244, or it may include tasks stored in both the mandatory tasks column 244 and the recommended tasks column 245. Furthermore, whether or not a task stored in the recommended tasks column 245 should be designated as a specific task may be determined based on information such as the skill level, work environment, and vehicle history stored in the user DB 22. For example, if the skill level is high, the work environment is adequate, or no recommended tasks have been performed on the vehicle within the most recent specified period, a task stored in the recommended tasks column 245 may be designated as a specific task.

[0050] Returning to the explanation of Figure 2, the work information 25 stores a first content containing multiple work information related to multiple tasks. Specifically, the work information 25 stores multiple first content, such as service manuals, instructions, recall information, parts information, and owner's manuals for each vehicle model provided by each manufacturer, reference information such as know-how held by the operator of system 5, and various information provided by external organizations such as government agencies and industry associations. The data format of the first content stored in the work information 25 can be any format, such as XML format, text format, PDF format, video format, or audio data format, and may differ for each first content.

[0051] Figure 5 illustrates a management table 25A that manages multiple first content items stored in work information 25. Management table 25A is included in work information 25, for example. In the example in Figure 5, the content of the first content, various tag information, and importance level are stored in association with each content ID stored in the content ID column 251.

[0052] The content field 252 stores the title of the first content. The content field 252 may also contain link information to the first content. The higher-level tag field 253 and the intermediate-level tag field 254 store tag information attached to the first content. In the example in Figure 5, the higher-level tag field 253 stores information about the manufacturer and vehicle model corresponding to the first content as tag information, and the intermediate-level tag field 254 stores information about the content of the first content as tag information. The importance level field 255 stores an index indicating the importance of the first content. Figure 5 is an example where the importance level is shown in five stages from "1" to "5", but there are no particular restrictions on how the importance level is expressed.

[0053] In the example in Figure 5, the first content labeled "AA001-2" is a modified version of the service manual labeled "AA001-1" by the manufacturer, and includes the entire service manual. Therefore, "AA001-2" has the intermediate tag "Modified (Full Text)". On the other hand, the first content labeled "AA002-2" is a modified version of the service manual labeled "AA002-1" by the manufacturer, showing only the modified portion (the difference from the original). Therefore, "AA002-2" has the intermediate tag "Modified (Difference)". In addition, both "AA001-2" and "AA002-2" have a higher importance level than the original first content.

[0054] Furthermore, in the example in Figure 5, the first content tagged "AA001-4" has the following intermediate tags: "rear brake pads," which are related to the content; "replacement," which is an attribute of the work related to the content; and "video," which indicates the format of the first content. In addition to the above example, various tags related to the content of the first content may be attached as tag information. Furthermore, tagging may be performed without hierarchical tagging. Also, there are no particular restrictions on the format of the tags; for example, HTML tags and tool control tags can be used.

[0055] The tag information and importance level of the first content are used, for example, in the important information extraction process described later. Various types of information, such as those stored in the upper tag field 253, the middle tag field 254, and the importance level field 255, are preferably directly attached to the first content itself as metadata. In this case, a management table 25A is not required, but it is preferable to provide a management table 25A from the viewpoint of improving the processing speed when extracting important information and reducing the workload of the processor 11.

[0056] Here, we will explain the first content with a specific example. Figure 6 is an example of the first content 300. The first content 300 is a service manual for vehicle model "AA001". The first content 300 was originally provided by manufacturer "AA", obtained by the operator of system 5, and stored in work information 25.

[0057] The first content section 300 includes "Repair Manual," "Instruction Manual," "Wiring Diagram," etc., as shown in tabs 301A to C. By selecting any of tabs 301A to C, the information displayed in the table of contents section 310 can be switched. Also, by selecting a table of contents item displayed in the table of contents section 310, the information displayed in the details section 320 can be switched. In the example in Figure 6, tab 301A is selected, and "Removal" of "Rear Brake Pads" in table of contents item 311 is selected.

[0058] The details section 320 displays multiple work information sections 321 to 324 related to the "removal" of the "rear brake pads" selected in the table of contents section 310. Work information section 321 shows the "work procedure". Work information section 321 includes text sections 321a and 321c that show each step in text, and image sections 321b and 321d that show illustrations or photographs related to each step.

[0059] Work information section 322 shows "post-work incidental tasks". Work information section 322 includes table 322a which summarizes "post-work incidental tasks". Work information section 323 shows "reference" information. Work information section 324 shows information that requires "caution". Work information section 324 includes text sections 324a and 324b which show each point of caution in text.

[0060] Furthermore, the first content 300 has tag information attached to it. The table of contents item 311 has a tag 311t attached to it, indicating that the table of contents item 311 is related to "rear brake pads". Although not shown in the illustration, in the example of Figure 6, other table of contents items are also tagged to indicate what the table of contents item is related to.

[0061] The work information fields 321 to 323 are tagged with tags 321t to 323t indicating the type and importance level of the work information. The text fields 324a and 324b within the work information field 324 are tagged with tags 324at and 324bt, respectively, indicating the type and importance level of the work information shown in the text. Tag information may be applied to the entire work information field 321, as in tag 321t, or to each piece of information contained within the work information field 324, as in tag 321at. In addition to the examples above, the tag information applied within the first content 300 may include tags indicating the difficulty level of the work, or other tags related to the content. The tag information applied within the first content 300 is used, for example, in the important information extraction process described later. As in these examples, tags are applied to at least a portion of the multiple pieces of work information contained within the first content. Preferably, the tags indicate, for example, the attributes of the work corresponding to the work information, and / or the importance level of the work.

[0062] Returning to the explanation of Figure 2, the first pre-trained model 26 is a pre-trained model capable of estimating important information in a task based on the attributes of the object being worked on (work object attributes) and / or the attributes of the task. The first pre-trained model 26 infers the content of the first content by performing, for example, natural language processing on the text contained in the first content, extracting predetermined keywords from the text, and analyzing predetermined tags attached to the first content. The first pre-trained model 26 also infers important information in the task based, for example, the result of this inference, the attributes of the object being worked on, and / or the attributes of the task.

[0063] Furthermore, this first pre-trained model 26 estimates important information by utilizing the correlation between various text data included in work information and the importance of the information indicated by that text data. In the field of vehicle maintenance, the following correlations are known as common technical knowledge. For example, text containing keywords such as "caution," "warning," and "important" contains important information related to safety and work quality. Also, in work procedures, text indicating prerequisites that affect subsequent work ("Be sure to check before..." or "Failure to do so will cause a malfunction") contains important information related to the success or failure of the work. And text included in recall information and modification information contains important information related to the latest safety standards and quality standards.

[0064] In this embodiment, the operator of System 5 trains the first trained model 26 using a training dataset created based on the correlation described above. The training dataset is, for example, work information text extracted from service manuals for multiple vehicle models, which has been labeled by maintenance professionals on a four-level scale: "important," "somewhat important," "reference," and "general." Note that other data, such as those exemplified below, may also be used as the training dataset.

[0065] The first pre-trained model 26 is created using, for example, a machine learning technique such as deep learning. The first pre-trained model 26 may be created using any method, such as supervised learning, unsupervised learning, semi-supervised learning, or a combination thereof. The first pre-trained model 26 can be created, for example, by training a dataset in which the data obtained by decomposing the first content into predetermined units (for example, item units such as parts information, work procedures, and specifications tables) is used as input parameters, and the important information contained in the data is used as output parameters. In other words, the first pre-trained model 26 can be created, for example, by machine learning using a dataset in which the input data is a predetermined text containing work information, and the ground truth data is information on the importance of various pieces of information that constitute the work information contained in the predetermined text. Which information is important can be determined, for example, by the operator of system 5. The first pre-trained model 26 may also be a rule-based pre-trained model that extracts important information from the first content by keyword pattern matching or the like.

[0066] Furthermore, the first pre-trained model 26 is composed of a neural network trained using, for example, the following training dataset. The training data consists of "work instruction texts" and "parts diagrams" extracted from a vast amount of past service manuals (first content) as input data, and "importance labels (e.g., 5-point scale)", "work type labels (e.g., replacement, inspection, removal / installation)", and "numerical information to focus on (e.g., torque value, wear limit value)" assigned by skilled mechanics or designers as ground truth data. In the training phase, the first pre-trained model 26 learns the statistical correlation (weight parameters) between specific words (e.g., "caution", "prohibition", "must"), contextual dependency relationships, and attributes of the corresponding object (vehicle type, year) contained in the input text and the ground truth data. As a result, even when unknown first content is input, the first pre-trained model 26 can infer and extract important information that the user should not overlook (e.g., the bolt tightening order for a specific vehicle type) with high accuracy, based on its context and the attributes of the object.

[0067] The fluctuation DB27 is a database that stores, for example, the content of tags attached to the first content and terms included in the first content that have the same meaning but differ in format or notation from manufacturer to manufacturer, by associating them with each other. The fluctuation DB27 is referenced, for example, when the first trained model 26 infers the content of the first content. The fluctuation DB27 may also be referenced, for example, when the second trained model 28 generates the second content. Having a fluctuation DB27 can reduce the impact of differences in format and notation from manufacturer to manufacturer.

[0068] The second pre-trained model 28 is a pre-trained model capable of generating second content using a predetermined generation method with important information as input. The second pre-trained model 28 is created, for example, using machine learning techniques such as deep learning. The second pre-trained model 28 may be created using any method, such as supervised learning, unsupervised learning, semi-supervised learning, or a combination thereof.

[0069] The second pre-trained model 28 can be created, for example, by training a dataset that uses important information contained in the first content as input parameters and the second content created based on that important information as output parameters. Alternatively, the second pre-trained model 28 may be created by fine-tuning a large, medium, or small general-purpose language model. Furthermore, the second pre-trained model 28 may be a rule-based pre-trained model that generates the second content according to the importance level assigned to the important information.

[0070] As a concrete implementation example of the second pre-trained model 28, a large-scale language model is used as the base model, and fine-tuning is performed specifically for generating service manual summaries. The input format includes important information (text) + generation method instruction (prompt) + user attributes (numerical values ​​and categories such as proficiency level), and the output format is the second content (HTML or Markdown format).

[0071] The training data for fine-tuning consists of the following elements: input data is important information extracted from a service manual; output data is various exemplary documents created by maintenance professionals (e.g., summary type, point-specific type, information-adding type, difference type, etc.); and metadata includes the target vehicle model, type of work, and the assumed skill level for each document. With this configuration, a person skilled in the art can generate practical second content using the second trained model 28 of this embodiment.

[0072] In this embodiment, the second trained model 28 includes a second trained model 28A and a second trained model 28B. The second trained model 28A is, for example, a trained model for generating second content for highly skilled users. The second trained model 28B is, for example, a trained model for generating second content for beginners with a low skill level. The second trained model 28A is a trained model that has been trained to include work information related to non-essential tasks stored in the recommended work column 245 in the second content, and to express important information in concise sentences using technical terms. The second trained model 28B is a trained model that has been trained to include only work information related to essential tasks in the second content, to express important information in easy-to-understand sentences without using technical terms, and to actively include videos, images, audio, etc. in the second content. The second trained model 28 may be, for example, different trained models for each other user attribute and / or for each object attribute.

[0073] The Item Management DB29A is a database that stores, for example, an object that is the target of a task indicated by the work information included in the first content, and item information relating to items that can be used in the task performed on that object. The Related Parts DB29B is a database that stores, for example, item information and information on related items that can be substituted for or used simultaneously with the item indicated by the item information. These will be described in detail later using Figure 13.

[0074] The control unit 30 is composed of, for example, a processor 11 and RAM. When the program 21 is executed by the processor 11, the control unit 30 functions as, for example, a user input acquisition unit 31, a user information acquisition unit 32, a task identification unit 33, an important information extraction unit 34, a supplementary information acquisition unit 35, a generation method determination unit 36, a second content generation unit 37, an output unit 38, an evaluation acquisition unit 39, an item extraction unit 40, a related item identification unit 41, a request unit 42, a first inventory management unit 43, a second inventory information acquisition unit 44, a request destination setting unit 45, a designation acceptance unit 46, a recording unit 47, and a task identification unit 48.

[0075] The user input acquisition unit 31 receives information entered by the user at the user terminal 2 and acquires it as user input information. The user information acquisition unit 32 acquires user information based on the user input information by referring to the user DB 22, which stores user information about the user.

[0076] The task identification unit 33 identifies a specific task from among multiple tasks related to multiple task information contained in the first content by referring to the first content DB 23, which stores a first content containing multiple task information related to multiple tasks, based on user information or user input information. User information or user input information may include, for example, quotation information that shows quotation items (such as work details and required items).

[0077] The task identification unit 33 identifies a specific task based, for example, on user information or user input information and the correspondence table 24. Alternatively, the task identification unit 33 may identify a specific task based, for example, on words or phrases contained in the user information or user input information. Specifically, the first trained model 26 may be used to identify a specific task from the task information 25 using these words or phrases as keywords.

[0078] The important information extraction unit 34 uses a first trained model 26 capable of estimating important information in a task based on the attributes of the object being worked on and / or the attributes of the task, to extract at least one piece of important information from at least one piece of task information related to the identified task.

[0079] The supplementary information acquisition unit 35 acquires supplementary information related to the extracted important information from a predetermined database or website provided by the external device 3, based on the important information extracted. The supplementary information acquisition unit 35 extracts supplementary information from related information 61, for example, using a first trained model. Alternatively, the supplementary information acquisition unit 35 may acquire supplementary information from related information 61 based on a keyword search using words and phrases contained in the important information, without using the first trained model.

[0080] The generation method determination unit 36 ​​determines the generation method for the second content based on a method determination element that includes at least one of the extracted important information, user attributes, and work attributes. The method determination element includes, for example, at least one of the user attributes and work attributes. Examples of user attributes used as a method determination element include at least one of the user's skill level, user's work history, and user's work environment. Examples of work attributes used as a method determination element include at least one of the work's importance level and work's difficulty level.

[0081] In addition, it is possible to specify user attributes and task attributes, and then determine the generation method based on these specified attributes. Furthermore, in addition to each attribute information, arbitrary supplementary information (surrounding information and background information) may be added, and the generation method for the second content may be determined based on this information. This allows for the generation of the second content using an appropriate method, and by using this second content, it becomes possible to perform more appropriate work by understanding the surrounding information and background information related to the information being generated.

[0082] Furthermore, "generation method" refers to the types of second content, such as the summary type, point-syntax type, information-adding type, and difference type mentioned above. The "generation method" may also be a type of second content determined according to the user's attributes, such as a beginner type that prioritizes readability and an expert type that prioritizes the amount of information contained in the second content. The "generation method" may also be a combination of the above types.

[0083] Furthermore, when generating second content using the second trained model 28, the "generation method" may be, for example, a prompt (instruction statement) that specifies each of the above types to the second trained model 28. Here, it is preferable that multiple instructions to be included in the prompt are stored in the storage unit 20. The generation method determination unit 36 ​​may create a prompt, for example, by obtaining instructions corresponding to method determination elements from the storage unit 20. The generation method determination unit 36 ​​may also change the instructions obtained from the storage unit 20 based on the user evaluation obtained by the evaluation acquisition unit 39. In addition, for example, if information about the language used by the user is obtained as a user attribute, the prompt may include an instruction to specify that language. In this case, the second content will be output in that language.

[0084] The second content generation unit 37 generates second content that includes at least a portion of the work information related to a specific task. This allows, for example, the system to appropriately extract and present to the user the information the user needs from content containing a large amount of work information, thereby reducing the burden on the user to perform a given task.

[0085] Furthermore, it is preferable that the second content generation unit 37 generates second content that includes at least a portion of the work information related to a specified task, based on work information related to a specific task and item information extracted by the item extraction unit 40 described later, and includes information that allows the user to request items corresponding to the extracted item information (for example, information that allows the user to view an order screen for ordering items). This makes it possible to appropriately extract and present the information that the user needs, while also reducing the burden of various request actions, such as ordering items such as parts and tools necessary for that task. As a result, the burden on the user for a given task can be further reduced.

[0086] Furthermore, it is preferable that the second content generation unit 37 generates second content related to the first content based on the extracted important information. Also, if supplementary information has been acquired by the supplementary information acquisition unit 35, it is preferable that the second content generation unit 37 generates second content based on, for example, at least the important information and the supplementary information.

[0087] The second content generation unit 37 preferably generates second content based on, for example, the extracted important information and the determined generation method. In this case, the second content generation unit 37 generates second content based on, for example, rules (conditions, specified information) predetermined for each generation method.

[0088] It is even more preferable that the second content generation unit 37 generates the second content using a second trained model 28 capable of generating the second content using a predetermined generation method with important information as input. In this case, for example, the extracted important information and the instruction sentence determined by the generation method determination unit 36 ​​are output to the second trained model 28 to generate the second content.

[0089] The second content generation unit 37 dynamically changes the DOM (Document Object Model) structure of the second content according to the user's skill level obtained by the user information acquisition unit 32. For example, if the skill level is "low," the second content generation unit 37 automatically adds "recommended spare parts" such as clips and bolts that are prone to breakage during the task, in addition to the extracted item information, by referring to the related parts DB 29B, and generates a list with checkboxes that can be ordered (dynamic control of second content generation). This reduces the risk of beginners interrupting their work due to broken parts.

[0090] The output unit 38 outputs the generated second content to the user terminal 2. Preferably, the output unit 38 outputs information to the user terminal in a manner that allows switching between the first content and the second content, or allows comparison between the first content and the second content.

[0091] The evaluation acquisition unit 39 acquires the user's evaluation of the second content via the user terminal 2. The evaluation acquisition unit 39 associates the second content with the user's evaluation of the second content and stores it in the storage unit 20. If the user's evaluation has been acquired by the evaluation acquisition unit 39, the generation method determination unit 36 ​​may further determine the generation method of the second content based on the acquired user evaluation.

[0092] For example, if the user ratings stored in the memory unit 20 meet predetermined conditions, the information processing device 1 updates the second content generation method. The predetermined conditions can be set as appropriate by the operator of the system 5, but examples include when the user ratings are lower than a predetermined threshold, or when the number of ratings exceeds a predetermined threshold.

[0093] The item extraction unit 40 extracts item information of items related to work information included in the first content. The extracted item information is used, for example, to enable the presentation of a request screen (e.g., an order screen) in the second content that requests the item.

[0094] The method for extracting items is not particularly limited, but it is preferable that the item extraction unit 40 extracts item information based on, for example, the item management DB 29A and multiple work information contained in the first content. In this case, for example, a pattern matching method may be used. With such a method, even if there are many first contents, for example, the parts that may be necessary for the work can be easily identified, and the risk of missing items can be reduced.

[0095] Furthermore, the item extraction unit 40 may extract item information using, for example, a trained model that infers and extracts items from the contents of the first content. Such a trained model can be created using machine learning techniques such as deep learning, and can be created using any method, such as supervised learning, unsupervised learning, semi-supervised learning, or a combination thereof. When performing supervised learning, for example, a dataset can be used as training data, in which a service manual that can become the first content is used as an input parameter, and items that can be used in the work indicated by the work information contained in the service manual are used as output parameters.

[0096] Furthermore, the item extraction unit 40 may extract item information relating to all items that can be used in all tasks indicated by all the work information included in the first content. In this configuration, it is preferable that the extracted item information is stored in the storage unit 20 in association with the first content. With this configuration, once item information has been extracted from the first content, it is not necessary to perform the item extraction process again, thus reducing the overall processing load of the system. In addition, the item extraction unit 40 may extract item information relating to items that can be used in tasks that have been identified as specific tasks from the work information included in the first content.

[0097] The item extraction unit 40 extracts noun phrases from the text data in the first content using natural language processing (morphological analysis, etc.) and matches them with the official name registered in the item management DB 29A or synonyms registered in the variation DB 27. Furthermore, the item extraction unit 40 may perform image recognition processing on the drawings (exploded views, etc.) in the first content to recognize the leader line numbers (part numbers) in the drawings and identify the item ID by comparing these with the bill of materials list. This prevents the omission of ordering parts that are shown in the drawings but not explicitly stated in the text.

[0098] The related item identification unit 41 identifies related items of the item indicated by the item information extracted by the item extraction unit 40, based on the related parts DB 29B. The identified related item information is used, for example, to enable the presentation of a request screen in the second content that requests the related item.

[0099] The request unit 42 requests an item based on receiving a predetermined operation input on the order screen from the user terminal 2. Here, information for displaying the request screen is output to the user terminal 2 by the output unit 38 in response to a predetermined operation being performed on the second content. In this embodiment, the case where "request" is "order" is mainly described, but "request" may be any of the various request actions described above. The term "order" in this specification can be read as any other request action.

[0100] The prescribed operation is not limited to, but could include, for example, selecting (clicking or tapping) link information that transitions to a request screen where items can be requested. The link information is set to, for example, the name, model number, or image of an item included in the content presented as second content. In this case, the second content includes the names, model numbers, or images of at least some of the items that can be used in the specified task. With such a configuration, the user can, for example, request the necessary parts with simple operations while confirming the content of the task to be performed.

[0101] Furthermore, the prescribed operation for displaying the request screen may be the selection of a button to initiate the request. In this case, for example, all items that can be used for a particular task may be displayed in a list, and the user may specify the requested quantity (e.g., order quantity). With such a configuration, for example, the efficiency of operations can be improved and the time required for various request actions can be reduced. Also, for example, the risk of missing item requests (e.g., missing orders) can be reduced.

[0102] The first inventory management unit 43 manages the first inventory information, which is inventory information for items on the user's side. Here, the second content includes, for example, information that enables the presentation of the first inventory information for items corresponding to extracted item information, and the output unit 38 preferably presents the first inventory information for items (such as the number of available items or the number of items that can be used) when outputting an order screen. With such a configuration, the user can easily determine, for example, whether or not to perform various request actions such as ordering for a certain item. Furthermore, the first inventory management unit 43 performs inventory management, such as increasing the allocated number of items or decreasing the inventory number, when, for example, the user selects to use an inventory item or actually uses it.

[0103] The second inventory information acquisition unit 44 acquires second inventory information, which is inventory information of the requesting party (e.g., a supplier). The second inventory information acquisition unit 44 acquires second inventory information by, for example, accessing the requesting party device 7. Here, it is preferable that the second content includes information that allows a list to be presented of, for example, the item corresponding to the extracted item information, the second inventory information of that item, related items identified by the related item identification unit 41, and the second inventory information of those related items. With such a configuration, the user can, for example, make appropriate decisions regarding various request actions for multiple candidate items, while considering the inventory status of the requesting party.

[0104] The request destination setting unit 45 sets at least one request destination based on user input information. The set request destination can be easily accessed or displayed preferentially on the request screen, for example. This configuration reduces the effort required for the user to search for and select a request destination, contributing to the speedup of various request-related tasks.

[0105] Furthermore, it is preferable that the request destination setting unit 45 sets at least one request destination for each attribute of the object and / or item, etc., that is the target of the work indicated by the work information included in the first content, based on user input information. It is also preferable that the request screen allows the user to select a request destination according to the attributes of the object and / or the attributes of the item to be ordered. With such a configuration, it becomes possible to efficiently select an appropriate request destination according to the attributes of the object and / or item, etc., and can contribute to the optimization and acceleration of work related to various request actions.

[0106] The designation reception unit 46 receives a designation from the user via the user terminal 2 for one or more tasks from among the tasks indicated by the task information contained in the second content. The designation reception unit 46 accepts the designation of a task based, for example, on an operation requesting to view the task details shown in the second content, or an operation selecting the task start button shown in the second content.

[0107] The recording unit 47 records the reception history related to the acceptance of work assignments. This configuration can, for example, support the management of work evidence based on information extracted as specific. The recording unit 47 may record, for example, the date and time the work assignment was accepted, the start date and time and end date and time of viewing the work content of the assigned work, or time information such as the cumulative viewing time. Furthermore, if there are multiple work contents shown in the second content, it is preferable for the recording unit 47 to record the reception history for each work content from the viewpoint of enabling detailed evidence management.

[0108] Furthermore, it is preferable that the recording unit 47 records, for example, a video of the user's work status in response to the designation reception unit 46 receiving a work designation, and associates it with the work information of the designated work. With such a configuration, reliable evidence can be provided in, for example, post-verification or audits. The video is captured by, for example, a camera 56 provided on the user terminal 2, a wearable camera as exemplified by the user terminal 2, or various other cameras. The videos captured by these cameras are transmitted to the information processing device 1 in real time, for example. The video captured by these cameras may also be configured to be transmitted to the information processing device 1 in response to the designation reception unit 46 receiving a work designation. In addition, the recording unit 47 may record still images generated from the video or still images acquired from the camera, instead of or in addition to the video.

[0109] Furthermore, the video may include footage captured by cameras installed in the user's workspace. Preferably, the video to be recorded is determined based on at least one of the following: information about the object being worked on, information about the user's workspace, and the user's work start time. This configuration allows the system to appropriately select the video to be recorded, even in environments where there are multiple workspaces or where cameras are constantly recording. In this configuration, the user DB22 includes at least one of the following: information about the object being worked on, information about the user's workspace, or the user's work start time.

[0110] Furthermore, the recording unit 47 acquires work completion information indicating that the user has performed work by referring to the second content, and stores history information in the user DB 22 based on the work completion information. The history information includes at least the date and time of work, the work location, information on the equipment used for the work, user identification information identifying the user who performed the work, and the specific details of the work. The date and time of work can be acquired from the system time of the user terminal 2 or the information processing device 1, or by user input. The work location can be acquired from the affiliation information stored in the user DB 22, from the location information (GPS, etc.) of the user terminal 2, or by user input. In addition, equipment information can be acquired from the work environment field 224 of the user DB 22, by reading the identification information (QR code (registered trademark), RFID, etc.) of the equipment used, or by user input. User identification information can be acquired from the user ID acquired at the time of login. The specific details of the work can be acquired from the work information included in the second content and the information on the work specified by the user.

[0111] One method for obtaining work completion information is to receive input from the user indicating work completion via user terminal 2. Alternatively, it is possible to acquire video or still images of the user's work status and automatically determine work completion based on these images. Furthermore, it is also possible to automatically determine work completion based on signals from sensors attached to the work object, or based on signals from tools or equipment used by the user (for example, a torque wrench tightening completion signal).

[0112] By storing this kind of historical information, the traceability of work is improved, making it possible to fully track who did what, when, where, and with what tools, thus facilitating quality control and post-work verification. In addition, it can support legal compliance, and in the field of automotive maintenance, it can be used to create maintenance records based on the Road Transport Vehicle Act. Furthermore, work efficiency is improved, and by referring to past historical information, past work performed on the same vehicle can be easily confirmed, contributing to increased efficiency and improved quality.

[0113] The task identification unit 48 identifies which task the task shown in the video corresponds to, based on the condition of the object being worked on and / or the user's posture. In this case, the recording unit 47 records the task information of the specified task and the video corresponding to that task, for example, based on the result of processing by the task identification unit 48. This configuration makes it easy to associate the specified task information with the video corresponding to that task, thereby improving the accuracy of evidence management. The above-mentioned reception history, videos, etc., are stored and managed in, for example, the user database 22.

[0114] The method for identifying the work situation is not particularly limited, but it is preferable to identify it based on at least one of the following: the shape of the work object included in the video, the user's work posture (bone data, etc.), and the relative positional relationship between the object and the user. The work identification unit 48 can, for example, take this information as features and use a trained model that has learned the correspondence between these features and work content to infer which work the work situation shown in the video corresponds to. This trained model may be, for example, a classification model based on deep learning. Alternatively, the work identification unit 48 may identify the work corresponding to the work situation shown in the video by using a pattern matching method on at least one of the following: the shape of the object, the user's work posture, and the relative positional relationship between the object and the user. These configurations can, for example, improve the accuracy of work situation identification. Furthermore, it becomes possible to, for example, determine in real time whether the user's work situation is in line with the work content to be performed and perform anomaly detection.

[0115] Here, we will describe in detail the training process of the pre-trained model (hereinafter referred to as the identification model) used in the work identification unit 48. The identification model is constructed using a training dataset that pairs previously recorded work video data with the correct labels (work ID, work name, correct / incorrect judgment result) assigned to the video data. Specifically, first, the regions of the worker and the object are identified from each frame of the video included in the training dataset, and the joint point coordinates (bone data) of the worker and the feature point coordinates of the object are extracted as time-series feature vectors.

[0116] Then, using this time-series feature vector as input and the corresponding ground truth label as output, the correlation between the two (for example, the probabilistic association between a sequence of joint movements in a specific joint and the label "screw tightening") is trained using algorithms suitable for time-series data processing, such as recurrent neural networks (RNNs) and transformers. As a result, when an unknown video is input, the discriminative model can infer the content of the work with high accuracy, not based on the pixel information in the image itself, but on abstract features such as the "type" of the worker's movements and their interaction with the object.

[0117] In connection with the processing by the recording unit 47, the output unit 38 outputs a list to the user terminal 2 or another computer device that presents, for example, the work information related to the work identified by the work identification unit 33 and the information recorded by the recording unit 47 in a comparable manner. This list can be used, for example, as a checklist for the user who performed the work or the user's manager to confirm that no work has been missed.

[0118] Furthermore, in connection with the processing by the recording unit 47, the output unit 38 outputs a third content to the user terminal 2 or another computer device, which presents, for example, work information included in the second content and a video recorded in association with the work information or a still image generated from the video. The third content can be used, for example, as a work report, an invoice to a customer, confirmation material for troubleshooting at a later date, or research material for work improvement. With this configuration, for example, the user's organization or customer can track work content with high accuracy, contributing to the improvement of business activities and increased customer satisfaction.

[0119] Furthermore, videos or still images in the third content may be presented, for example, as link information that allows access to the videos or still images. From the viewpoint of facilitating access to videos, etc., it is preferable that the third content includes, for example, a QR code (registered trademark) for accessing videos or still images.

[0120] Here, if the user input information or user information includes quotation information that shows the quotation items, the work identification unit 33 may identify a specific work from among multiple works based on the quotation information. In this case, if an invoice is output as third content, for example, the customer can compare the quotation items in the quotation with the actual work performed, with evidence such as videos, which can further improve customer satisfaction.

[0121] The user terminal 2 comprises a storage unit 50, a control unit 52, a display device 55, and a camera 56. The storage unit 50 is composed of, for example, a storage device such as a ROM, HDD, or SSD included in the user terminal 2. The storage unit 50 stores a program 51. The program 51 is, for example, an information processing program for implementing various functions described later in the user terminal 2.

[0122] The control unit 52 is composed of, for example, a processor and RAM included in the user terminal 2. When program 51 is executed by the processor, the control unit 52 functions as, for example, a reception unit 53 and a presentation unit 54.

[0123] The reception unit 53 receives input information from the user, for example, via a predetermined input interface such as a keyboard or touchscreen. The input information received by the reception unit 53 is transmitted to the information processing device 1 as, for example, user input information.

[0124] The presentation unit 54 presents the second content to the user, for example via the display device 55, based on information about the second content received from the information processing device 1. The display device 55 is, for example, a display that shows the second content. The display device 55 may be a touchscreen. The display device 55 may be built into the user terminal 2 or it may be an external device that can be connected to the user terminal 2. The camera 56 is a camera capable of shooting video. At least a portion of the video captured by the camera 56 is transmitted to the information processing device 1, for example, and recorded in the user DB 22, etc., by the recording unit 47.

[0125] Here, "display" is shown as an example of presenting the second content, but it is not limited to this. As shown above, the second content may also be audio data, so "audio output" could also be used as an example of posting. Note that this audio data is generated by recognizing the content of the generated second content and converting that text into audio.

[0126] In particular, in the case of voice output (text-to-speech), instead of outputting all the characters recognized in the second content, it is also possible to output only the voice of specific character information (specific information) from all the character information. In this case, the specific information is only the important information extracted through the extraction process from the information represented by the second content, which is output as voice.

[0127] Furthermore, when comparing and outputting, the audio of the first and second contents may be output alternately. In this case, it is desirable to output the type of content being output before outputting the content of each.

[0128] System 5 may be implemented by including a native application downloaded to the user terminal 2, by including a web application, or by being implemented as a cloud-based service such as SaaS.

[0129] The external device 3 comprises a storage unit 60 and a control unit 62. The storage unit 60 is composed of, for example, a storage device such as a ROM, HDD, or SSD included in the external device 3. The storage unit 60 stores related information 61 related to the first content. The control unit 62 is composed of, for example, a processor or RAM included in the external device 3. The control unit 62 transmits the related information 61 to the information processing device 1 in response to receiving a request from the information processing device 1 to refer to the related information 61.

[0130] The requesting device 7 comprises a storage unit 70 and a control unit 73. The storage unit 70 is composed of, for example, a storage device such as a ROM, HDD, or SSD included in the requesting device 7. The storage unit 70 stores a program 71 and second inventory information 72. When the program 71 is executed by the processor, the control unit 73, for example, receives a request from the information processing device 1 to refer to the second inventory information 72 and transmits the second inventory information 72 to the information processing device 1. The control unit 73 also, for example, receives a request from the information processing device 1 regarding various request actions and starts processing to ship the requested items, etc. The requesting device 7 may be included in the system 5. In this case, for example, the second inventory information 72 may be managed in the storage unit 20 of the information processing device 1. The information processing device 1 may also execute at least a part of the processing to ship the requested items, etc.

[0131] Next, we will describe the information processing method performed by System 5. Figure 7 is a flowchart schematically showing an example of the information processing method according to the embodiment of this disclosure. The order of the various processes that constitute the various flows described below may be in any order and may be executed in parallel, as long as no inconsistencies occur in the processing content.

[0132] As shown in Figure 7, the information processing method in this embodiment executes the second content-related processing (step S10), request support processing (step S20), and evidence trail support processing (step S30). For the sake of explanation, these processes are a classification of the series of steps in this embodiment according to their main purpose. The second content-related processing is a series of processes whose main purpose is to generate the second content. The request support processing is a series of processes whose main purpose is to support requests for items, etc., related to the second content. The evidence trail support processing is a process whose main purpose is to support the management of evidence trails related to the work shown in the second content. In the following, we will explain using the example where "specific work" is "work that the user should perform" and "request" is "ordering," but "specific work" and "request" are not limited to these examples.

[0133] Furthermore, the processing results of the steps included in each process can be used in other processes. Also, if the configuration is designed to achieve only a specific main objective, there may be steps that can be omitted in other processes separate from the process related to that main objective. In addition, the processing order of the steps included in each process can be in any order among the three processes, as long as no inconsistencies occur in the processing content, and they may be executed in parallel.

[0134] First, let's explain the second content-related processing. Figure 8 is a flowchart showing an example of the second content-related processing shown in Figure 7. In step S101, user terminal 2 performs login processing. Known methods may be used for login processing. In login processing, for example, the user ID and login password entered into user terminal 2 are sent to information processing device 1 as user input information. Information processing device 1 compares the received user input information with information stored in user DB 22, etc., and performs login authentication.

[0135] In step S102, the user terminal 2 receives user input. In step S102, the user inputs information such as the task the user intends to perform, details of the task, and vehicle attributes such as the vehicle model name or chassis number of the vehicle to which the task is performed. The input format may be, for example, a selection input format in which the user selects the appropriate item from a set of items, or a free input format. The user terminal 2 transmits the information entered by the user as user input information to the information processing device 1.

[0136] In step S103, the information processing device 1 obtains user input information transmitted from the user terminal 2 in step S102. In step S104, the information processing device 1 obtains user information. Specifically, the information processing device 1 obtains user information from the user DB 22 based on the user ID obtained during the login process in step S101. The information processing device 1 may also update the user DB 22 based on the vehicle attribute information obtained in step S103.

[0137] In step S105, the information processing device 1 identifies the task that the user should perform. In step S105, the task is identified based, for example, on the user input information, including the task the user intends to perform, the details of the task, and information about the vehicle's attributes, and the correspondence table 24. Whether or not the recommended task stored in the recommended task column 245 should be the task the user should perform may be determined, for example, based on the user's skill level and vehicle history stored in the user database 22.

[0138] If the determination is based on the user's skill level, for example, it is possible to classify any given task as a different task (similar task) even if it is related to that task. For example, if the user's skill level is "high," tasks that involve disassembling and repairing, including removing parts, may be identified. If the user's skill level is "medium," tasks that involve disassembling and repairing without removing parts when repairable may be identified, while tasks that involve replacing parts without repairing them when removal is necessary may also be identified. Furthermore, if the user's skill level is "low," tasks that involve replacing parts may be identified.

[0139] If input was accepted in a free-form input format in step S102, for example, the user's task may be identified based on the result of matching the text received via free-form input with the keywords contained in the keyword field 246.

[0140] In step S106, the information processing device 1 obtains keywords. In step S106, for example, it obtains keywords associated with the tasks that the user should perform by referring to the keyword column 246 of the correspondence table 24. In step S106, the name of the tasks that the user should perform may also be obtained as a keyword.

[0141] In step S107, the information processing device 1 performs an extraction process for important information. Specifically, the information processing device 1 uses the first trained model 26 to extract important information that is important for the task that the user should perform from among at least one piece of work information contained in each of the multiple first contents stored in the work information 25.

[0142] Here, with reference to Figure 9, the process of extracting important information in step S107 will be described in detail. Figure 9 is a flowchart showing an example of the process of extracting important information shown in Figure 8.

[0143] If the first content is tagged (YES in step S121), in step S122, the trained model 26 analyzes the tags attached to the first content. Tag analysis is performed on, for example, the tags attached to the first content itself and the tags attached to the work information, etc., contained in the first content. Tag analysis may also be performed on the tag information contained in the management table 25A. If the first content is not tagged (NO in step S121), the process proceeds to step S124.

[0144] If the content of the first content can be inferred based on the analyzed tags (YES in step S123), proceed to step S125. If the content of the first content cannot be inferred based on the analyzed tags (NO in step S123), in step S124, the trained model 26 performs natural language processing and / or keyword extraction.

[0145] Natural language processing can be performed in any way, but examples include morphological analysis, syntactic analysis, semantic analysis, and contextual analysis. Keyword extraction is a process that extracts keywords obtained in step S106 from the first content, for example, using methods such as pattern matching.

[0146] In step S125, the trained model 26 infers the content of the first content based on the tags analyzed in step S122 and / or the results of the processing performed in step S124. In step S124, the model infers the work information contained in the first content, such as the object (e.g., vehicle type), object details (e.g., brakes, etc.), work type (e.g., replacement, etc.), work content (e.g., brake pad replacement, etc.), and / or work details (e.g., tire removal and installation during brake pad replacement, brake pad replacement, rotation check, etc.).

[0147] In step S126, the trained model 26 estimates important information. Based on, for example, the inference result in step S125, the attributes of the vehicle that the user is supposed to work on, and / or the attributes of the work, the trained model 26 estimates which of the multiple pieces of work information are important information for the work the user is supposed to work on.

[0148] In step S126, important information may be estimated based on keywords obtained in step S106, for example. Also in step S126, if the first content has been modified, the modified portion may be estimated to be important information. The modified portion can be extracted, for example, based on the first content before modification and the first content after modification.

[0149] In step S127, the trained model 26 extracts important information from the first content based on the estimation results from step S126. After step S127, the process of extracting important information is completed. Then, the process proceeds to step S108 shown in Figure 8.

[0150] Returning to the diagram description, in step S108, the information processing device 1 creates a summary of the first content based on the extracted important information. In step S108, for example, the first trained model 26 is used to create the summary so that it contains at least one important piece of information.

[0151] In step S109, the information processing device 1 acquires supplementary information. In step S109, for example, the first trained model 26 is used to infer information contained in the related information 61, and supplementary information is acquired based on the inference result and the important information extracted in step S108. Alternatively, in step S109, similar to step S126, supplementary information may be acquired based on the inference result for the related information 61 and the attributes of the vehicle that is the target of the work to be performed by the user, and / or the attributes of the work.

[0152] The contents of related information 61 and the supplementary information obtained in step S109 may be stored as new first content in work information 25, etc. If the trained model 26 determines that related information 61 does not contain any information that should be used as supplementary information, no supplementary information will be obtained.

[0153] In step S110, the information processing device 1 determines the method for generating the second content. In step S110, for example, a prompt to be output to the second trained model 28 is determined as the generation method. The prompt includes, for example, the important information extracted in step S107 and the supplementary information acquired in step S109. The prompt also includes, for example, information specifying the type of the second content. Preferably, the type of the second content is determined based on at least one of the following: the user's skill level, the user's work history, the importance level of the work, and the difficulty level of the work.

[0154] If the user's skill level is low, the prompts may include instructions such as, "Create in simple language," "Highlight essential points," "Provide explanations of technical terms," ​​"Provide links to related information 61," or "Provide videos if available."

[0155] In step S111, the information processing device 1 generates second content. Specifically, the second trained model 28 generates second content based on the prompt determined in step S110. The second trained model 28 used in step S111 may be changed according to the user's attributes and / or the attributes of the object. For example, the second trained model 28 used in step S111 may be changed according to the user's skill level, whether it is an engine-powered vehicle or an electric vehicle, etc. With such a configuration, the content of the second content can be further tailored to the user.

[0156] In step S112, the information processing device 1 outputs the second content. Specifically, the information processing device 1 transmits information about the second content generated in step S111 to the user terminal 2. The user terminal 2 receives the information about the second content.

[0157] In step S113, the user terminal 2 presents the second content based on the information received in step S112. In step S113, for example, the second content is displayed on the display device 55.

[0158] In step S114, the user terminal 2 receives the user's evaluation of the presented second content. The user's evaluation may be received, for example, in a selection input format, in a free input format, or in a combination of both.

[0159] Here, we will explain the second content with a specific example. Figure 10 is an example of the second content 400A displayed on user terminal 2. The second content 400A is created based on the first content 300 shown in Figure 6. The second content 400A is a summary type.

[0160] Similar to the first content 300, the second content 400A includes "Repair Manual," "Instruction Manual," "Wiring Diagram," etc., as shown in tabs 401A to C. By selecting any of tabs 401A to C, the information displayed in the table of contents section 410 can be switched. Also, by selecting a table of contents item displayed in the table of contents section 410, the information displayed in the details section 420 can be switched.

[0161] The details section 420 displays multiple pieces of work information related to the "removal" of the "rear brake pads" selected in the table of contents section 410. The work information sections 421, 422, 424, text sections 421a, 421c, 424a, 424b, image sections 421b, 421d, and table 422a in the second content 400A correspond to the work information sections 321, 322, 324, text sections 321a, 321c, 324a, 324b, image sections 321b, 321d, and table 322a in the first content 300, respectively.

[0162] The second content 400A consists only of information related to tasks that the user should perform and that has been extracted as important information. In other words, the table of contents items in the table of contents section 410 of the second content 400A are a part of the table of contents items in the table of contents section 310 of the first content 300. Also, the task information in the details section 420 of the second content 400A is a part of the task information in the details section 320 of the first content 300.

[0163] On the user terminal 2, virtual buttons 431 to 433 are displayed in addition to the second content 400A. Virtual button 431 is a button for changing the display format of the second content. When a tap operation is performed on virtual button 431, the type of the second content is changed. When a tap operation is performed on virtual button 431, the user can select a generation method for the second content, such as summary type, point highlighting type, information addition type, difference type, beginner type, and expert type. It is preferable that multiple types can be selected.

[0164] When the virtual button 431 is selected and the type of second content is selected, information regarding the type selected by the user is transmitted from the user terminal 2 to the information processing device 1. The information processing device 1 performs the same processing as in steps S110 to S112, for example, in order to generate the type of second content desired by the user and present it to the user. Alternatively, all types of second content may be generated in advance in steps S110 and S111, and information for presenting all types of second content may be transmitted to the user terminal 2 in step S112.

[0165] The virtual button 432 is a button for inputting an evaluation of the second content 400A. When the virtual button 432 is tapped or otherwise operated, for example, a screen for receiving an evaluation of the second content 400A is displayed on the user terminal 2.

[0166] The virtual button 433 is a button for displaying the first content 300, which is the basis for the second content 400A. When the virtual button 433 is tapped or otherwise operated, for example, the first content 300 may be displayed instead of the second content 400A, or the first content 300 and the second content 400A may be displayed side by side for comparison. In this comparative display, it is desirable to clearly indicate the type of each content (First Content 300 and Second Content 400A). By clearly indicating the type of content in this way, it becomes clear that First Content 300 is the original, legitimate content, and Second Content 400A is related content generated based on First Content. This clarifies the hierarchical relationship between the content based on their types. This clarifies that the creator of First Content bears ultimate responsibility for its content, while Second Content is merely used supplementarily (to improve work efficiency and productivity) and therefore bears no responsibility.

[0167] Figure 11 illustrates a second content 400B generated using a different generation method than the example in Figure 10. The second content 400B shown in Figure 11 is an example of the second content 400A shown in Figure 10 being modified into a type that combines summary type, point-specification type, and information-added type.

[0168] Compared to the second content 400A, the second content 400B includes point information 441-445 and additional information 446. Point information 441-445 is important information extracted from a first content different from the first content 300. Point information 441-445 is information that the second trained model 28 has determined to be of a particularly high importance level among the important information extracted in step S107.

[0169] Additional information 446 is supplementary information obtained from related information 61. Additional information 446 contains a link to a video. When additional information 446 is selected, for example, video and audio related to the work procedure described in text field 421c are played.

[0170] Returning to the explanation of Figure 8, in step S115, the information processing device 1 obtains the user's evaluation of the second content. In step S116, the information processing device 1 associates the second content with the user's evaluation and stores it in the storage unit 20.

[0171] If the user evaluation stored in the memory unit 20 satisfies a predetermined condition (YES in step S117), in step S118, the information processing device 1 updates the second content generation method. In step S118, for example, when generating prompts for the same user or other users based on the same or similar method determination elements, the information processing device 1 updates the selection priority of various instructions so that at least one instruction included in the prompt is different. After that, the information processing device 1 terminates the series of processes included in step S10. If the user evaluation stored in the memory unit 20 does not satisfy a predetermined condition (NO in step S117), the information processing device 1 terminates the series of processes included in step S10.

[0172] Next, the request support process will be explained. Figure 12 is a flowchart showing an example of the request support process shown in Figure 7. In step S201, the information processing device 1 extracts item information from the first content. In step S201, for example, the item information is extracted by pattern matching the item name, model number, or image contained in the item management DB 29A with the strings and images described in the multiple work information contained in the first content. The target of pattern matching may be the entire first content, or it may be only the part that describes the work that the user is supposed to perform, as identified in step S105 above.

[0173] Step S201 may be performed, for example, between steps S105 to S112 in Figure 8, or simultaneously with any of these steps. Alternatively, step S201 may be performed in advance at a predetermined timing, such as when the first content is registered in the first content DB. In this case, the item information extracted from the first content is stored in the storage unit 20 in association with the first content, for example.

[0174] In step S202, the information processing device 1 identifies related items. In step S202, for example, the information processing device 1 refers to the related parts DB29B to identify related items associated with the items extracted in step S201.

[0175] Figure 13 illustrates the information stored in the item management DB and the related parts DB. In the example in Figure 13, information about items that can be used in the object identified by the vehicle name 291 and model 292, which are information used to identify the object that may be the target of the work, is comprehensively stored as item information 293. Item information 293 includes, for example, the item name, model number, and item ID. Item information 293 may also include image information of the item or other information. By storing multiple pieces of information for identifying an item in item information 293, the possibility of missing items during pattern matching in step S201 can be reduced.

[0176] Furthermore, information on related items is comprehensively stored as related item information 294, associated with each item shown in item information 293. Related item information 294 includes, for example, the name, model number, item ID, and type of the related item. The type indicates, for example, a substitute for that item, or a tool used simultaneously with that item. In the case of a substitute, for example, the type of substitute such as a second-hand genuine part or a superior part may be stored in the type column.

[0177] As shown in the example in Figure 13, by comprehensively storing information on items that can be used with a particular object, it becomes possible to easily extract items from each of the corresponding first content items, even if there are multiple types of first content items that correspond to that object. The item management DB29A and related parts DB29B are updated as needed, for example, when the manufacturer of the object updates information about the object.

[0178] Returning to the explanation of Figure 12, in step S203, the information processing device 1 sets a link for transitioning to the order screen in the second content generated in step S111 of Figure 8, based on the extracted item information and the identified related item information. The transition link is set, for example, to the name, model number, or image of an item included in the content presented as the second content.

[0179] In step S204, the information processing device 1 receives an order processing request from the user via the user terminal 2. The user makes an order processing request, for example, by performing a predetermined operation on the second content. The predetermined operation is, for example, selecting a transition link or selecting the order start button.

[0180] In step S205, the information processing device 1 obtains first inventory information, which is information about the inventory owned by the user, by referring to the user DB 22 for the items extracted in step S201 and the related items identified in step S202.

[0181] In step S206, the information processing device 1 outputs an order screen to the user terminal 2. If a transition link set in the name of an item in the second content was selected in step S204, the order screen becomes, for example, a screen for ordering that item and related items of that item.

[0182] Figure 14 is a schematic diagram showing an example of an order screen. Specifically, Figure 14 is an example of a display screen output to the user terminal 2 in step S206. In Figure 14, the order screen 510 is displayed in a pop-up format from the second content 400B. The second content 400B also has transition links 501A to 501C set up. The transition links 501A to 501C are set to a string indicating the item name, a string referring to the item's part number (model number), and an image of the item, respectively. When a transition link 501A to 501C is selected, the order screen for the corresponding item is displayed.

[0183] In the example in Figure 14, the order screen 510 is a window that appears when the transition link 501A is selected. Therefore, on the order screen 510, the item to be ordered is "Auxiliary Battery Terminal," and the item ID of this item is displayed. The order screen 510 also displays the inventory quantity based on the first inventory information obtained in step S205.

[0184] Furthermore, virtual buttons 511 to 513 are displayed on the order screen 510. When virtual button 511 is selected, inventory management processing is executed, such as increasing the allocated quantity of the target item, and no order is placed. When virtual button 511 is selected, for example, the suppliers set in step S207 described later are extracted and displayed. When virtual button 513 is selected, for example, a screen is displayed that lists the selectable suppliers, or a screen is displayed for placing an order using a bidding method.

[0185] In addition, in the example in Figure 14, a virtual button may be provided to display a list of all items that can be used in the work shown in the second content 400B, and to accept orders, in addition to or instead of the transition links 501A to 501C.

[0186] Returning to the explanation of Figure 12, in step S207, the information processing device 1 sets the supplier based on user input. The process in step S207 may be executed at any time desired by the user, and may be executed in advance, for example, before step S101 in Figure 8.

[0187] Supplier settings can be configured, for example, from the user's My Page. Furthermore, suppliers can be set based on attributes of the object or item, such as by manufacturer, type, or part type. For example, a car parts supplier might have a wide selection of parts from a particular manufacturer but not from others, or they might specialize in the selection and pricing of specific parts. By allowing suppliers to be set based on attributes of the object or item, it becomes possible to select the most suitable supplier for each attribute, making it easier to obtain items and reducing procurement costs.

[0188] Furthermore, it may be possible to set a priority order for suppliers, with higher priority suppliers appearing higher in the list used for selecting suppliers. It may also be possible to set priorities for each attribute of the object or for each item.

[0189] In step S208, the information processing device 1 accepts the user's selection of an ordering method via the user terminal 2. Any ordering method may be adopted, and for example, the system operator can set it as appropriate. In this embodiment, we will explain using as examples an ordering method in which a supplier is specified (when the virtual button 512 in Figure 14 is selected) and an ordering method using a bidding system without specifying a supplier (when the virtual button 513 in Figure 14 is selected). Note that in step S208, an option to use the company's own inventory without placing an order may be provided, as shown by the virtual button 511 in Figure 14.

[0190] If specifying a supplier is selected in step S208, in step S209, the information processing device 1 extracts and outputs the suppliers set in step S207. If suppliers are set for each attribute such as the object or item, in step S209, suppliers corresponding to those attributes are extracted. If a priority order for suppliers is set, for example, they are displayed on the user terminal 2 in the order according to the priority order.

[0191] In step S210, the information processing device 1 receives the selection of a supplier from the user via the user terminal 2. In step S211, the information processing device 1 obtains second inventory information from the request destination device 7 or the storage unit 20. In step S212, the information processing device 1 outputs a list screen of items to be ordered to the user terminal 2. In step S213, the information processing device 1 receives the selection of order details from the user via the user terminal 2.

[0192] If the logged-in user has the authority to order items (YES in step S214), in step S215, the information processing device 1 receives an order confirmation operation from the user via the user terminal 2 and processes the order based on the order details received in step S213. If the logged-in user does not have the authority to order items (NO in step S214), the information processing device 1 sends an approval request to a person with ordering authority. Once approval is granted based on the approval request, step S215 described above is executed.

[0193] Here, we will describe in detail the processing in steps S209 to S215 using Figures 15 to 17. Figures 15 to 17 are schematic diagrams showing examples of order screens. Specifically, Figure 15 shows the order screen 520 output to the user terminal 2 in step S209. At the top of the order screen 520, there is an order target field 521 that displays information such as the items to be ordered. Below the order target field 521, the suppliers extracted in step S209 are displayed. In the example in Figure 15, when virtual button 522A is selected, "Parts Supplier XX" is selected as the supplier, and when virtual button 522B is selected, "Parts Supplier YY" is selected as the supplier.

[0194] On the order screen 520, the system may be configured to allow the order target to be changed, for example, if a string indicating the target product is selected, or if a virtual button (not shown) for changing the order target is selected. When the order target is changed, for example, suppliers corresponding to that order target are extracted again, and the extracted suppliers are displayed on the order screen 520 so that they can be selected. On the order screen 520, a virtual button (not shown) may be provided to call up a list of other suppliers not set in step S209.

[0195] Figure 16 specifically shows the order screen 530 output to the user terminal 2 in step S212. The order screen 530 is also the display screen after parts supplier XX is selected in Figure 15.

[0196] At the top of the order screen 530, there is an order target field 531 that displays information about the items to be ordered. Below the order target field 531, there is a first order acceptance field 532 that displays the price, type, and stock quantity at the supplier for the items to be ordered and related items that can be used as substitutes for those items. In step S213, the order details are accepted by accepting selections, for example, by selecting a virtual button 532A displayed in the order field of the first order acceptance field 532 or a virtual button 532B displayed in the restock field. The selected items are then added to the user's shopping cart, for example.

[0197] Furthermore, below the first order acceptance field 532, there is a field for accepting orders for related parts, which are items that can be used simultaneously with the item being ordered. In this way, the order screen 530 is configured to allow ordering of related items in addition to the item being ordered.

[0198] Furthermore, in the example shown in Figure 15, it may be possible to select multiple suppliers, and the order screen 530 may display information such as the price and availability of items from the multiple suppliers selected in Figure 15 in a comparative manner. In addition, the second inventory information and price may be configured to be displayed on the order screen 520 in Figure 15.

[0199] Figure 17 specifically shows the order screen 540 that is output to the user terminal 2 in step S214. The order screen 530 is the screen that appears after the order details have been specified in Figure 16, and is the shopping cart screen that is accessed in response to a predetermined operation by the user.

[0200] At the top of the order screen 540, there are tabs 541 for displaying the shopping cart screen and tab 542 for displaying the user's My Page. The order screen 540 displays the information selected by tab 541. If tab 542 is selected, for example, it becomes possible to view or change user information.

[0201] The cart column 543 displays the order details received in step S213. If the user has the authority to place orders, selecting the virtual button 544 completes the order process with the contents shown in the cart column 543. If the user does not have the authority to place orders, for example, as shown in Figure 17, the virtual button 544 will be displayed in a grayed-out manner and will not be selectable. In this case, selecting the virtual button 545 sends an approval request to the authorized person via system 5. If the authorized person approves, the system may be configured to complete the order process immediately, or the user who requested the approval may be notified that the order has been approved, and the order process may be completed by accepting the selection of the virtual button 544 from the user.

[0202] Returning to the explanation of Figure 12, if the bidding method is selected in step S208, in step S217, the information processing device 1 receives the order conditions from the user via the user terminal 2. The order conditions include, for example, the amount, delivery deadline, and bidding deadline.

[0203] If a bid meeting the order requirements is received within the bidding deadline (YES in step S218), the process proceeds to step S215 and the order processing is completed. If no bid meeting the order requirements is received within the bidding deadline (NO in step S218), the process returns to step S208, and, for example, the selection of the order method is accepted again.

[0204] Figure 18 is a schematic diagram showing an example of an order screen. The order screen 550 shown in Figure 18 is an example of a display screen output to the user terminal 2 in step S217. The order screen 550 displays an order target field 551 that displays information such as the items to be ordered, and an order conditions field 552 that displays the order conditions. If, for example, a string indicating the order conditions is selected in the order conditions field 552, input becomes possible to change the order conditions. In the example in Figure 18, a pull-down menu 552A for changing the part type is shown. By specifying the order conditions and selecting the virtual button 553, these order conditions are notified to multiple suppliers.

[0205] Order screen 560 is the screen displayed when a bid is received within the bidding deadline. In this case, the order process is completed by automatic ordering. Order screen 570 is the screen displayed when no bids are received within the bidding deadline. In this case, if virtual button 571 is selected, the process returns to step S208, and if virtual button 572 is selected, the series of order processes is canceled.

[0206] Next, we will explain the evidence support process. Figure 19 is a flowchart of an example of the request support process shown in Figure 7. In step S301, the information processing device 1 receives a specification of the work shown in the second content from the user via the user terminal 2. In step S302, the information processing device 1 records the history of the received work specification.

[0207] Figures 20 and 21 are schematic diagrams showing examples of task specification acceptance screens, respectively. Specifically, Figure 20 is an example of a display screen output to the user terminal 2 in step S301. In Figure 20, a virtual button 601 is provided at the bottom of the second content 400B. The virtual button 601 is the button selected when starting the task shown in the second content 400B. In step S301, for example, when the virtual button 601 is selected by the user, the specification of the task corresponding to the virtual button 601 (in the example of Figure 20, removal of rear brake pads in maintenance) is accepted.

[0208] Furthermore, if the virtual button 601 is selected, step S302 records, for example, the date and time the request was received and the end date and time the user viewed the work information related to the requested work as the reception history. The end of viewing can be determined, for example, when the user transitions from the page displaying the work information to another page. Alternatively, the user's gaze and facial orientation may be used to determine whether or not they actually viewed the work information and the duration of that viewing. The reception history is stored, for example, in association with the requested work information and managed as evidence of the work.

[0209] Furthermore, when the user selects the virtual button 602 located at the top of the second content 400B, the corresponding task is added to the work items. The user can also select multiple work items to be performed and then work on multiple work items together.

[0210] The display screen 610 shown in Figure 21 is, for example, the screen used when a user selects multiple work items. In the display screen 610, two work items have been selected by the user. When the virtual button 611 is selected in this state, other tasks included in the second content 400B become available for selection as work items.

[0211] When the user selects the virtual button 612 to start work on display screen 610, a second content is displayed, for example, containing work information for multiple work items shown on display screen 610. In display screen 620 (the screen transitioned from display screen 610) shown in Figure 21, the second content first displays work information 621 for removing the rear brake pads. Work information 621 corresponds to work procedure "1." shown in Figure 20, and the text field 421a and image field 421b are also based on the content shown in Figure 20. When the user selects the virtual button 623 in this state, content corresponding to work procedure "2." shown in Figure 20 is displayed, for example. The virtual button 622 is a button to return to the previous screen.

[0212] In step S301, for example, when the virtual button 612 is selected, it may be determined that the entire task, "remove the rear brake pads," has been specified. In this case, a second content that shows the entire task in a list format may also be displayed. By configuring it in this way, for example, evidence that the user has viewed the second content can be secured, while reducing the number of times the user operates the user terminal 2, thereby reducing the burden on the user.

[0213] Furthermore, in step S301, for example, it may be determined that the work procedure "1." has been specified when the virtual button 612 is selected. In this case, for example, the reception history can be managed for each of the multiple work procedures included in the work item "removal of rear brake pads," making it possible to secure more detailed and accurate evidence.

[0214] Similarly, each time a virtual button to proceed to the next screen is selected, the content of a task item, such as "removal of rear brake pads," divided according to a predetermined division criterion, will be displayed on screen 620. The division criterion is not particularly restricted and can be determined as appropriate by the operator of System 5.

[0215] Furthermore, the content displayed on the display screen 620 may include, for example, the point information 441-443 and notes 445 that were included in the second content 400B. In this case as well, from the viewpoint of ensuring high-precision evidence, it is preferable to record the reception history for point information 441, etc., using the method described above. Furthermore, even if the virtual button 601 for starting work is selected in the example of Figure 20, the system may be configured to sequentially display the content included in the work item, divided according to a predetermined division criterion, in response to user operation, as shown in Figure 21.

[0216] Returning to the explanation of Figure 19, in step S303, the information processing device 1 acquires the video to be recorded. The acquisition of the video is performed, for example, in response to the user's specification of a task in step S301. The video is captured, for example, by a camera installed in the user terminal 2 or a wearable camera worn by the user, in response to the acceptance of the task specification, and transmitted to the information processing device 1 in real time.

[0217] Furthermore, if the workplace is constantly being imaged by a stationary monitoring camera installed in the user's workspace, in step S303, for example, the case information 22B is referred to, and the video to be recorded is identified and acquired from among multiple videos. The video captured by the monitoring camera may be configured to be transmitted to the information processing device 1 at a predetermined timing. From the viewpoint of being able to detect user omissions or dangerous situations for the user in real time, it is preferable to configure the system to transmit the captured video to the information processing device 1 in real time.

[0218] Figure 22 is a diagram illustrating the information stored in the user database 22. Specifically, Figure 22 illustrates case information 22B that may be referenced in step S303. In the example of Figure 22, associated with each case ID stored in the case ID field 231, the vehicle type, user ID, work item, work date and time, work details, browsing history, and video link corresponding to each case ID are stored.

[0219] The vehicle type field 232 stores information about the vehicle type to be worked on in that case. The user ID field 233 stores the ID of the user in charge of the work in that case. The work item field 234 stores the work items that are scheduled to be performed in that case. The work location field 235 stores information indicating the work location for that case. In this example, it is assumed that there are multiple work locations. The work details field 236 stores information indicating the details of the work that are scheduled to be performed in that case. The work details are, for example, information obtained by dividing the multiple work items shown in the work item according to a predetermined division criterion. The division criterion is not particularly limited and can be determined as appropriate by the operator of system 5. In the example in Figure 22, the division criterion is set based on the inspection items recorded in the maintenance record book that records the contents and history of statutory inspections, etc.

[0220] In step S303, for example, based on information such as the vehicle type, work date and time, and work location stored in association with the case ID, the memory unit 20 identifies a video from among multiple videos stored in the memory unit 20 that is presumed to capture the work status of that case.

[0221] The browsing history field 237 and the video link field 238 contain information for purposes such as audit trail management. The browsing history field 237 stores information such as whether the user has viewed the second content showing the work details for each work detail, that is, whether the user has requested the work in step S301. The video link field 238 stores link information to the video recorded in step S307, which will be described later. In the example in Figure 22, the user with user ID "001-1" is in the middle of working on the case with case ID "C01". Therefore, browsing history and video links are stored for work details that have been completed after the work was requested, but browsing history and other information are not stored for work details that have not yet been started.

[0222] Returning to the explanation of Figure 19, in step S304, the information processing device 1 identifies which task the work situation shown in the video corresponds to. In step S304, for example, the device identifies which task the work situation shown in the video corresponds to by using an inference model or pattern matching method as described above.

[0223] If the task specified by the user in step S301 matches the user's task identified in step S304 (YES in step S305), the process proceeds to step S307. If they do not match (NO in step S305), in step S306, the information processing device 1 performs error processing.

[0224] If the process proceeds to step S306, it is highly likely that the user's work status does not conform to the work that should be performed. Therefore, in step S306, as an error handling measure, for example, an alert is sent to user terminal 2 instructing it to review its work. Even if the process proceeds to step S306, it is preferable that the video recording of the work is stored in association with the work in order to enable post-verification.

[0225] In step S307, the information processing device 1 stores in the user DB 22 the work information of the task specified by the user in step S301 and the video identified as matching the task and its content, in association with each other. Preferably, this work information and video are associated with a case ID, as shown in Figure 22.

[0226] In step S308, the information processing device 1 outputs a checklist to the user terminal 2 or another computer device that presents, in a comparable manner, the work information related to the work identified in step S105 of Figure 8 and the evidence information recorded in steps S302 and / or S307, depending on whether predetermined conditions have been met. The predetermined conditions are not particularly limited, but include, for example, receiving an output request from a device such as the user terminal 2, or determining that the work identified in step S105 has been completed.

[0227] Figure 23 is a schematic diagram showing an example of a checklist. In the checklist 630 shown in Figure 23, a case information display area 631 is provided in the upper left. Below the case information display area 631, a check sheet 632 is provided. The check sheet 632 includes a work location area 633, an inspection item area 634, a target area 635, a browsing history area 636, and a video link area 637.

[0228] In the example in Figure 23, the work location column 633 shows the work locations for the work items. The inspection item column 634 shows the items to be inspected at each work location. The target column 635 has checks next to the inspection items that should be performed in this work, i.e., the inspection items that correspond to the work that the user identified in step S105 should perform. Inspection items that are not checked in the target column 635 do not need to be performed in this work. Note that the checklist may only include the inspection items that should be performed in this work.

[0229] The browsing history field 636, similar to the browsing history field 237 described above, indicates whether the user viewed the second content showing the work details for each inspection item, that is, whether the work details were specified in step S301. The video link field 637, similar to the video link field 238 described above, contains link information to videos recorded in step S307 in association with each inspection item. Making it possible to output such a checklist makes it easier for administrators and other mechanics to check it, and it can also be used as evidence in the event of an audit by a third party.

[0230] In this embodiment, instead of using existing service manuals as reference materials for the user, a second content dynamically generated from existing service manuals, etc., is used. Therefore, it is difficult to prepare a checklist of work procedures in advance. However, by making it possible to output a checklist as shown in Figure 23, even when using dynamically generated service manuals, etc., it becomes possible to reduce the risk of missing work checks, increase the reliability of checks, and achieve efficient evidence management.

[0231] Returning to the explanation of Figure 19, in step S309, the information processing device 1 outputs a third content to the user terminal 2 or other computer device in response to a request from the user terminal 2 or other computer device. This third content presents the work information contained in the second content in association with a video recorded in association with the work information or a still image generated from the video. In step S310, the information processing device 1 outputs the video or still image to the requesting terminal in response to a request to access a link contained in the third content, and then terminates.

[0232] Figure 24 is a schematic diagram illustrating an example of the third content. In Figure 24, an invoice 640 is shown as an example of the third content. The invoice 640 includes an invoice section 641 that lists the billing details and amount, and a details section 642 that shows the details of the work performed by the user. The details section 642 includes, for example, codes 643A and 643B, which are QR codes (registered trademarks) containing details of the work and links to videos or still images of the work being performed. Customers who receive the invoice can scan codes 643A and 643B with a smartphone camera to view images of the work performed by the user. In the example in Figure 24, a code containing link information to an image is included for each work area, but these codes may be different for each inspection item.

[0233] Furthermore, from the perspective of improving customer satisfaction, it is preferable to obtain quotation information showing the quotation items in step S103 or step S104 of Figure 8, and to identify the tasks that the user should perform in step S105 based on these quotation items. Through such processing, for example, the quotation items presented to the customer can be clearly identified as tasks that the user should perform, and transparency regarding the work content can be ensured for the customer.

[0234] Although one embodiment of the present invention has been described in detail above, the present invention is not limited to the above-described embodiment, and modifications and improvements can be made as appropriate. The present invention includes all modifications within the meaning and scope of the claims and equivalents thereof.

[0235] <Note 1> This disclosure provides an information processing system that identifies tasks that users should perform from a service manual (first content) for vehicle maintenance, extracts the necessary item information for those tasks, and generates and provides a second content that integrates with an ordering function. The information processing apparatus comprises at least one processor and a storage device for storing an information processing program, wherein the processor executes the information processing program to perform: a user input acquisition process for acquiring user input information from a user terminal used by a user; a user information acquisition process for acquiring user information based on the user input information by referring to a user database for storing user information relating to the user; a task identification process for identifying a task to be performed by the user from among the multiple tasks related to the multiple task information included in the first content, based on the user information or the user input information, by referring to a first content database for storing a plurality of task information each related to a plurality of tasks; and an item relating to an item that can be used in the task indicated by the task information included in the first content. It is possible to perform an item extraction process for extracting item information; a second content generation process for generating second content that includes at least a part of the work information relating to the identified work and information that enables the display of a screen requesting an item corresponding to the extracted item information, based on the work information relating to the work identified by the work identification process and the item information extracted by the item extraction process; an output process for outputting the second content to the user terminal; a work completion acquisition process for acquiring work completion information indicating that the user has performed the identified work by referring to the second content; and a history storage process for storing history information in the user database based on the work completion information, including at least the date and time of the work, the work location, the equipment information used for the work, user identification information identifying the user who performed the work, and the specific details of the work.

[0236] <Note 2> This disclosure provides a system that uses AI to extract important information from a vast amount of service manuals (first content) in the field of vehicle maintenance and repair, and dynamically generates customized second content according to the user's skill level and work environment. The information processing device comprises a processor and a memory connected to the processor, the memory storing a first trained model and a second trained model as programs, and the processor, by executing the programs, can refer to a first content database that stores first content including work information relating to work on an object, refer to a user database that stores user attributes including user proficiency, work history and work environment, and perform important information estimation processing using the first trained model to estimate important information from the first content based on work attributes and object attributes, generation method determination processing that determines a generation method for second content using the second trained model based on user attributes, content generation processing that generates second content from the estimated important information according to the determined generation method using the second trained model, and output processing that outputs the second content to a user terminal. [Explanation of Symbols]

[0237] 1: Information Processing Device 2, 2A, 2B, 2C: User terminals 3: External device 5: System 7: Request destination device

Claims

1. An information processing apparatus comprising at least one processor and a memory device for storing an information processing program, The processor executes the information processing program, A user input acquisition process that obtains user input information from the user terminal used by the user, A user information acquisition process that acquires the user information based on the user input information by referring to a user database that stores user information about the user, A task identification process that identifies a specific task from among the multiple tasks related to the multiple task information contained in the first content, based on the user information or user input information, by referring to a first content database that stores a first content containing multiple task information related to multiple tasks, each of which is associated with multiple tasks, An item extraction process that extracts item information of items related to the work information included in the first content, A second content generation process generates a second content that includes at least a portion of the work information relating to the identified work, and which includes information that makes it possible to request an item corresponding to the extracted item information, based on work information relating to a specific work and extracted item information. Output processing for outputting the second content to the user terminal, An information processing device that performs this task.

2. The item extraction process includes an item management database that stores item information relating to an object that is the target of work indicated by the work information included in the first content and an item that can be used in work on said object, and a process of extracting the item information based on the plurality of work information included in the first content. The information processing apparatus according to claim 1.

3. The aforementioned information processing device further, The system executes an inventory management process that manages the first inventory information, which is the inventory information of the item on the user side. The second content includes information that enables the presentation of first inventory information for items corresponding to the extracted item information. The information processing apparatus according to claim 1 or 2.

4. The aforementioned information processing device further, Based on a related parts database that stores the aforementioned item information and related items that can be substituted for or used simultaneously with the item indicated by the aforementioned item information in association with each other, a related item identification process is performed to identify the related items of the item indicated by the extracted item information. The second content includes information that enables the presentation of a request screen in which the related items can be requested. The information processing apparatus according to claim 1 or 2.

5. The aforementioned information processing device further, A second inventory information acquisition process is executed to acquire second inventory information, which is the inventory information of the requested item and related items. The second content includes information that enables a list presentation of items corresponding to the extracted item information, second inventory information for said items, related items identified by the related item identification process, and second inventory information for said related items. The information processing apparatus according to claim 4.

6. The aforementioned second content includes the names, model numbers, or images of at least some of the items that may be used in the identified work, The name, model number, or image presented as the second content contains link information that leads to a request screen where the item can be requested. The information processing apparatus according to claim 1 or 2.

7. The aforementioned information processing device further, Based on the user input information, a request destination setting process is executed to set at least one request destination for each attribute of the object that is the target of the work indicated by the work information included in the first content, and / or for each attribute of the item. In the request screen where the aforementioned item can be requested, the recipient can be selected according to the attributes of the object and / or the attributes of the item. The information processing apparatus according to claim 1 or 2.

8. The aforementioned information processing device further, Using a first trained model created by machine learning with a dataset in which predetermined text containing work information is input data and information on importance attached to various pieces of information constituting the work information contained in the predetermined text is used to perform an important information extraction process that extracts at least one important piece of information from at least one piece of work information related to the identified work. The second content generation process generates the second content such that it includes the extracted important information. The information processing apparatus according to claim 1 or 2.

9. An information processing method performed by an information processing apparatus comprising at least one processor and a memory device for storing an information processing program, In the aforementioned information processing device, A user input acquisition step that obtains user input information from the user terminal used by the user, A user information acquisition step involves obtaining user information based on user input information by referring to a user database that stores user information about the aforementioned user, A task identification step involves referring to a first content database that stores a first content containing a plurality of task information related to a plurality of tasks, and identifying a specific task from among the plurality of tasks related to the plurality of task information contained in the first content, based on the user information or the user input information. An item extraction step for extracting item information of items related to work information included in the first content, A second content generation step of generating a second content that includes at least a portion of the work information relating to the identified work, based on work information relating to a specific work and extracted item information, and includes information that makes it possible to request an item corresponding to the extracted item information. Output step of outputting the second content to the user terminal, An information processing method that includes performing the following.

10. An information processing program executed in an information processing device comprising at least one processor and a memory device for storing an information processing program, When the information processing program is executed by the processor, the information processing device will: A user input acquisition process that obtains user input information from the user terminal used by the user, A user information acquisition process that acquires the user information based on the user input information by referring to a user database that stores user information about the user, A task identification process that identifies a specific task from among the multiple tasks related to the multiple task information contained in the first content, based on the user information or user input information, by referring to a first content database that stores a first content containing multiple task information related to multiple tasks, each of which is associated with multiple tasks, An item extraction process that extracts item information of items related to the work information included in the first content, A second content generation process generates a second content that includes at least a portion of the work information relating to the identified work, and which includes information that makes it possible to request an item corresponding to the extracted item information, based on work information relating to a specific work and extracted item information. Output processing for outputting the second content to the user terminal, An information processing program that executes [something].

11. An information processing system comprising at least one computer, The aforementioned at least one computer, A user input acquisition process that obtains user input information from the user terminal used by the user, A user information acquisition process that acquires the user information based on the user input information by referring to a user database that stores user information about the user, A task identification process that identifies a specific task from among the multiple tasks related to the multiple task information contained in the first content, based on the user information or user input information, by referring to a first content database that stores a first content containing multiple task information related to multiple tasks, each of which is associated with multiple tasks, An item extraction process that extracts item information of items related to the work information included in the first content, A second content generation process generates a second content that includes at least a portion of the work information relating to the identified work, and which includes information that makes it possible to request an item corresponding to the extracted item information, based on work information relating to a specific work and extracted item information. Output processing for outputting the second content to the user terminal, An information processing system that performs [this action].

Images