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

The system addresses the inefficiency of manual material preparation by using worker attributes and language models to generate tailored work procedures, enhancing operational efficiency.

JP2026062018AActive Publication Date: 2026-04-09DAIKIN INDUSTRIES LTD +1
View PDF 5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2026-04-09

AI Technical Summary

Technical Problem

Existing systems require manual preparation of work materials based on worker proficiency levels, leading to significant operation man-hours.

Method used

An information processing system that generates work procedure information automatically using a computer that acquires worker attributes and employs language models to tailor procedures based on skill level, qualifications, and other attributes.

Benefits of technology

Automatically generates customized work procedure information, reducing man-hours and improving efficiency by providing appropriate guidance for workers based on their attributes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026062018000001_ABST
    Figure 2026062018000001_ABST
Patent Text Reader

Abstract

Providing information processing methods, information processing devices, information processing systems, and computer programs. [Solution] The system acquires attribute information indicating the attributes of the worker performing the work, and uses one or more language models that refer to the acquired attribute information and standard procedure information indicating the standard procedure for the work to generate work procedure information according to the worker's attributes, and then a computer executes a process to output the generated work procedure information.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a technique of storing a work manual including predetermined work contents and work procedures and worker information regarding a worker, and switching a presentation method regarding the work contents and work procedures according to the proficiency level of the worker.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In Patent Document 1, it is necessary to assume in advance what information should be presented according to the proficiency level of the worker and store it in the system. Therefore, it is necessary to prepare materials in advance according to the proficiency level of the worker, and there is a problem that the operation man-hour is large.

[0005] An object of the present disclosure is to provide an information processing method, an information processing apparatus, an information processing system, and a computer program that can automatically generate work procedure information according to the attributes of a worker.

Means for Solving the Problems

[0006] The information processing method relating to the first aspect of this disclosure involves a computer that acquires attribute information indicating the attributes of a worker performing a task, generates work procedure information corresponding to the worker's attributes using one or more language models that refer to the acquired attribute information and standard procedure information indicating the standard procedure for the task, and outputs the generated work procedure information.

[0007] The information processing method relating to the second aspect of this disclosure, in the information processing method relating to the first aspect, wherein the attribute information includes at least one of the worker's skill level, qualifications, work speed, and evaluation of the worker.

[0008] The information processing method relating to the third aspect of this disclosure is an information processing method relating to the second aspect, wherein the evaluation includes at least one of the following: the rate of frequent service, the rate of free replacement of multiple parts, the rate of service becoming paid, the average number of visits, and the rate of inquiries to the call center.

[0009] The information processing method relating to the fourth aspect of this disclosure further acquires at least one of the following in an information processing method relating to any one of the first to third aspects: information relating to the work, information relating to the equipment to be worked on, information relating to past cases in the work, and supplementary information relating to the standard procedure; and the computer performs a process to generate work procedure information corresponding to the attributes of the worker using the attribute information, the standard procedure information, and a language model that references the acquired information.

[0010] The information processing method relating to the fifth aspect of this disclosure, in the information processing method relating to the fourth aspect, includes at least one of the following: information about the work site and information about the time constraints in the work.

[0011] The information processing method relating to the sixth aspect of this disclosure, in the information processing method relating to the fourth aspect, includes at least one of the following regarding the equipment: configuration information of the equipment, measured values ​​measured for the equipment, control values ​​input to the equipment, and intended use of the equipment.

[0012] The information processing method relating to the seventh aspect of this disclosure is an information processing method relating to any one of the first to sixth aspects, wherein the computer acquires at least one of imaging data captured at the work site and acoustic data recorded at the work site, and uses the attribute information, the standard procedure information, and a language model that references the acquired at least one of the data to generate work procedure information corresponding to the attributes of the worker.

[0013] The information processing method relating to the eighth aspect of this disclosure is an information processing method relating to any one of the first to seventh aspects, wherein the computer receives a request from an worker for work procedure information and performs a process to generate work procedure information corresponding to the received worker request using the language model.

[0014] The information processing device relating to the ninth aspect of this disclosure comprises at least one control unit, the control unit acquires attribute information indicating the attributes of a worker performing a task, generates work procedure information corresponding to the worker's attributes using one or more language models that refer to the acquired attribute information and standard procedure information indicating the standard procedure for the task, and outputs the generated work procedure information.

[0015] The information processing system relating to the tenth aspect of this disclosure comprises a worker's terminal device and a server device that is communicably connected to the terminal device and has a storage unit that stores standard procedure information indicating the standard procedure for a task. The server device acquires attribute information indicating the attributes of the worker performing the task, and uses one or more language models that refer to the acquired attribute information and the standard procedure information stored in the storage unit to generate work procedure information corresponding to the attributes of the worker performing the task, and outputs the generated work procedure information.

[0016] A computer program according to the 11th aspect of the present disclosure causes a computer to execute a process of acquiring attribute information indicating an attribute of an operator who performs a task, generating task procedure information according to the attribute of the operator using one or more language models that refer to the acquired attribute information and standard procedure information indicating the standard procedure of the task, and outputting the generated task procedure information.

Advantages of the Invention

[0017] According to the present disclosure, task procedure information can be automatically generated according to the attributes of an operator.

Brief Description of the Drawings

[0018] [Figure 1] It is an explanatory diagram for explaining an outline of a process executed by an information processing system according to an embodiment. [Figure 2] It is a block diagram showing an internal configuration of a wearable device. [Figure 3] It is a block diagram showing an internal configuration of a server device. [Figure 4] It is a conceptual diagram showing an example of a task management table. [Figure 5] It is a conceptual diagram showing an example of an attribute database. [Figure 6] It is a conceptual diagram showing an example of a standard procedure database. [Figure 7] It is an explanatory diagram for explaining an example of generation of a revised procedure manual. [Figure 8] It is a schematic diagram showing an example of a prompt input to a language model. [Figure 9] It is a flowchart for explaining a procedure of a process executed by a server device.

Modes for Carrying Out the Invention

[0019] Hereinafter, an information processing system according to an embodiment will be specifically described based on the drawings. Figure 1 is an explanatory diagram illustrating the outline of the processing performed by the information processing system 1 according to the embodiment. The information processing system 1 according to the embodiment comprises a wearable device 10 worn by a worker and a server device 20 that manages the work process. The wearable device 10 and the server device 20 are connected to each other so as to be able to communicate via a communication network NW such as the Internet. In addition, an LLM server 30 is connected to the communication network NW, and the server device 20 is able to use the language model MD installed in the LLM server 30 via the communication network NW.

[0020] In this embodiment, the worker is, for example, a worker who performs installation, repair, and inspection work related to refrigeration and air conditioning equipment. Refrigeration and air conditioning equipment refers to equipment connected to refrigerant piping and constituting a refrigeration cycle. Refrigeration and air conditioning equipment includes equipment such as outdoor units, indoor units, ventilation systems, heat source equipment, and central control devices that constitute an air conditioning system, and equipment such as refrigerators, heat exchangers, and control devices that constitute a refrigeration system.

[0021] The server device 20 generates work procedure manuals (work procedure information) according to the worker's attributes by utilizing the language model MD. Here, worker attributes include at least one of the worker's skill level, qualifications, work speed, and evaluation of the worker. The work procedure manual is data that defines the work procedures for tasks such as installation work, repair work, and inspection work, and consists of multiple work procedures. Standard work procedures for each task are assumed to be created in advance by the work manager, etc., and registered in the standard procedure database DB2 (see Figure 6). The server device 20 transmits the worker's attribute information and standard procedure information indicating the standard procedure for the work performed by the worker to the LLM server 30, and inputs them into the language model MD to generate work procedure manuals according to the worker's attributes.

[0022] The Language Model (MD) outputs a revised work procedure manual (hereinafter also referred to as the revised procedure manual) based on the worker's attributes, when the worker's attribute information and the standard procedure information for the work performed by the worker are input. For example, if the worker has a high level of skill, the Language Model (MD) may output a revised procedure manual that omits or simplifies the explanation of the work procedure, or a revised procedure manual that skips some of the work procedure. Conversely, if the worker has a low level of skill, the Language Model (MD) may output a revised procedure manual that adds or elaborates on the explanation of the work procedure, or a revised procedure manual that adds some preliminary work procedures.

[0023] The server device 20 transmits the revised procedure manuals for each worker, generated using the language model MD, to the wearable device 10 via the communication network NW.

[0024] The wearable device 10 is a device having imaging, recording, audio output, and communication functions. The wearable device 10 is worn, for example, around the neck of a worker. Alternatively, the wearable device 10 may be worn on the worker's head, or on another body part such as the shoulder or arm. Furthermore, the wearable device 10 may be a goggle-type camera device, or any device with the above functions, such as a smartphone or action camera, may be used instead of the wearable device 10.

[0025] The wearable device 10 obtains revised procedure manuals from the server device 20 regarding the work to be performed by the worker. The wearable device 10 provides voice guidance according to the revised procedure manual obtained from the server device 20. For example, the wearable device 10 can refer to the revised procedure manual and provide voice instructions on the work that the worker should perform. The wearable device 10 may also refer to the revised procedure manual and provide voice inquiries about the work that the worker has performed. The worker performs the work according to the voice guidance output from the wearable device 10, confirms the work that has been performed, and responds to inquiries from the wearable device 10 by voice.

[0026] The wearable device 10 starts capturing images and recording audio when it outputs voice guidance corresponding to each work procedure. Since the wearable device 10 is worn by the worker, the images obtained by the wearable device 10 are from the worker's perspective (i.e., first-person perspective). In this embodiment, the images obtained by the wearable device 10 are videos captured at an appropriate frame rate, but they may also be still images captured at appropriate time intervals. The images captured by the wearable device 10 include the work scene, the equipment being worked on, and various measuring instruments. The sounds recorded by the wearable device 10 include the worker's voice, sounds emitted from the equipment being worked on, ambient sounds, and any other sounds generated at the work site. It is preferable that the sounds recorded by the wearable device 10 include the voice guidance mentioned above.

[0027] The wearable device 10 terminates imaging and recording when a work procedure is completed. In this embodiment, imaging data and acoustic data are obtained for each work procedure. The imaging data and acoustic data for each work procedure are examples of work data in this disclosure.

[0028] The work data may include imaging data and acoustic data for each work procedure, as well as work result data entered by the worker. The work results are entered into the wearable device 10 as voice responses to inquiries from the wearable device 10. The work result data may be voice data or text data converted by speech recognition.

[0029] The work data may include imaging data and acoustic data for each work procedure, as well as equipment data obtained for the equipment being worked on. The equipment data includes measured values ​​such as current, voltage, electrical resistance, power, temperature, airflow, and wind speed measured for the equipment being worked on. The measured values ​​may be values ​​measured by measuring instruments or sensors installed in the equipment being worked on, or values ​​measured by an operator using measuring instruments or sensors. The equipment data may also include status data and error codes output from the equipment being worked on.

[0030] The wearable device 10 uploads the acquired work data to the server device 20 via the communication network NW.

[0031] The server device 20 may determine whether the work performed by the worker is appropriate based on the work result data uploaded from the wearable device 10. If the server device 20 determines that the work performed by the worker is appropriate, it can give instructions to the wearable device 10 to prompt the worker to move on to the next work procedure. On the other hand, if the server device 20 determines that the work is not appropriate, it can give instructions to the wearable device 10 and have it output voice guidance prompting the worker to correct the work, thereby guiding the worker to perform the work appropriately.

[0032] In addition to the wearable device 10, the worker may also carry a mobile device such as a smartphone. The mobile device has an input unit such as a touch panel and a display unit such as an LCD display, and complements some of the functions that the wearable device 10 does not have. The mobile device is configured to communicate with the wearable device 10 using short-range wireless communication such as Bluetooth® or ZigBee®. The mobile device is also configured to communicate with the server device 20 via a communication network NW.

[0033] Figure 2 is a block diagram showing the internal configuration of the wearable device 10. The wearable device 10 includes a control unit 11, a storage unit 12, a communication unit 13, an imaging unit 14, an audio input unit 15, an audio output unit 16, a sensor unit 17, and the like.

[0034] The control unit 11 includes a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and the like. The ROM in the control unit 11 stores control programs and the like that control the operation of each hardware component of the wearable device 10. The CPU in the control unit 11 reads and executes the control programs and the like stored in the ROM, and controls the operation of the hardware components, thereby making the entire device function as the terminal device of this disclosure. The RAM in the control unit 11 temporarily stores data used during the execution of various processes.

[0035] The control unit 11 may also include functions such as a clock for outputting date and time information, a timer for measuring the elapsed time from the time a measurement start instruction is given until a measurement end instruction is given, and a counter for counting numbers.

[0036] The storage unit 12 is equipped with an auxiliary storage device and stores various types of data, as well as computer programs executed by the control unit 11. The data stored in the storage unit 12 includes work data (imaging data, acoustic data, processing result data, and equipment data) acquired for each work procedure. The computer programs stored in the storage unit 12 include a transmission control program that transmits the work data acquired for each work procedure to the server device 20.

[0037] The communication unit 13 is equipped with a communication module for wireless communication with external devices such as the server device 20. The communication module uses a communication module for wireless communication using known mobile communication standards such as 3G, 4G, and 5G, or wireless LAN methods such as WiFi (registered trademark). The communication unit 13 communicates with external devices such as the server device 20 via the communication network NW, transmits necessary data such as work data, and receives appropriate data transmitted from the external devices.

[0038] The communication unit 13 may be equipped with a communication module for short-range wireless communication, such as Bluetooth® or ZigBee®, in order to communicate with a mobile terminal such as a smartphone carried by the worker.

[0039] The imaging unit 14 includes an optical lens, an image sensor, a driver circuit, and the like. A wide-angle lens is preferably used as the optical lens. The image sensor is a CMOS (Complementary Metal Oxide Semiconductor), CCD (Charge-Coupled Device), etc., and generates electrical signals according to the intensity of light formed through the optical lens. The driver circuit includes a timing generator (TG), etc., and sequentially reads electrical signals from the image sensor in synchronization with the clock signal output from the TG to generate video data. The video data generated by the imaging unit 14 is sent to the control unit 11 and stored in the storage unit 12. The video data generated by the imaging unit 14 is also transmitted to the server device 20 via the communication unit 13.

[0040] The sound input unit 15 includes a microphone for collecting sound, a processing circuit for converting the collected sound into a digital signal (acoustic data), and the like. The acoustic data generated in the sound input unit 15 is sent to the control unit 11, where it undergoes appropriate processing such as noise reduction, and is then stored in the storage unit 12. The acoustic data processed by the control unit 11 is then transmitted to the server device 20 via the communication unit 13.

[0041] The sound output unit 16 is equipped with a speaker that outputs sound. The sound output unit 16 outputs sound based on acoustic data provided by the control unit 11.

[0042] The sensor unit 17 is equipped with non-contact sensors for detecting the worker's fingers and other body parts. The sensors in the sensor unit 17 include proximity sensors and gesture sensors. For example, the proximity sensor detects when the worker's fingers come within a predetermined range. For example, the gesture sensor detects the movement of the worker's fingers. The detection results from the sensor unit 17 are notified to the control unit 11. The control unit 11 may control the operation of the wearable device 10 based on the detection results from the sensor unit 17.

[0043] Figure 3 is a block diagram showing the internal configuration of the server device 20. The server device 20 is a dedicated or general-purpose server device and includes, for example, a control unit 21, a storage unit 22, a communication unit 23, an operation unit 24, and a display unit 25.

[0044] The control unit 21 includes a CPU, ROM, RAM, etc. The ROM in the control unit 21 stores control programs and the like that control the operation of each hardware component of the server device 20. The CPU in the control unit 21 reads and executes the control programs stored in the ROM and the computer programs described later stored in the memory unit 22, and by performing processes to control the operation of the hardware components, the entire device functions as the server device 20 of this disclosure. The RAM in the control unit 21 temporarily stores data used during the execution of various processes.

[0045] In this embodiment, the control unit 21 is configured to include a CPU, ROM, and RAM, but the configuration of the control unit 21 is not limited to the above. The control unit 21 may be one or more processing circuits, for example, including a GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), quantum processor, volatile or non-volatile memory, etc. The control unit 21 may also include functions such as a clock that outputs date and time information, a timer that measures the elapsed time from the time a measurement start instruction is given to the time a measurement end instruction is given, and a counter that counts numbers.

[0046] The storage unit 22 is equipped with a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The storage unit 22 stores various computer programs executed by the control unit 21, and various data used when the computer programs are executed. The computer programs stored in the storage unit 22 include a procedure generation program PG that causes the computer to execute a process that acquires attribute information indicating the attributes of the worker performing the work, generates work procedure information according to the worker's attributes using a language model MD that references the acquired attribute information and standard procedure information indicating the standard procedure for the work, and outputs the generated work procedure information.

[0047] A computer program including a procedure generation program PG may be provided on a non-temporary recording medium RM on which the computer program is recorded in a readable format. The recording medium RM is a portable memory such as a CD-ROM, USB memory, SD card, microSD card, or CompactFlash®. The control unit 21 reads various computer programs from the recording medium RM using a reading device (not shown in the figure) and stores the read computer programs in the storage unit 22. The computer programs stored in the storage unit 22 may also be provided via communication. In this case, the control unit 21 acquires the computer programs via communication through the communication unit 23 and stores the acquired computer programs in the storage unit 22.

[0048] The procedure generation program PG may be a single computer program or may consist of multiple computer programs. The procedure generation program PG may be executed on a single computer or may be executed collaboratively by multiple computers (e.g., server device 20 and LLM server 30). Furthermore, the procedure generation program PG may partially utilize existing libraries.

[0049] Furthermore, the storage unit 22 includes a work management table TB, an attribute database DB1, and a standard procedure database DB2. The contents stored in the work management table TB, attribute database DB1, and standard procedure database DB2 will be explained in detail using Figures 4 to 6.

[0050] The communication unit 23 includes a communication module for wireless communication with external devices such as the wearable device 10. The communication module may utilize known mobile communication standards such as 3G, 4G, and 5G, or wireless LAN methods such as Wi-Fi (registered trademark). The communication unit 23 communicates with external devices such as the wearable device 10 via a communication network NW, receiving data such as field-of-view images (video data) and transmitting appropriate data to the external device. The communication unit 23 may further include a communication module for short-range wireless communication such as Bluetooth (registered trademark) or ZigBee (registered trademark).

[0051] The operation unit 24 is equipped with operating devices such as a touch panel, keyboard, and switches, and accepts various inputs and operations from work managers and others. The control unit 21 acquires information input through the operation unit 24 and performs appropriate control based on the various operation information provided by the operation unit 24.

[0052] The display unit 25 is equipped with a display device such as a liquid crystal monitor or an organic EL (Electro-Luminescence) monitor, and displays information that should be notified to the work manager, etc., in response to instructions from the control unit 21.

[0053] The server device 20 may be a single computer, or it may be a computer system composed of multiple computers and peripheral devices. Furthermore, the server device 20 may be a virtualized virtual machine, or it may be a cloud.

[0054] Figure 4 is a conceptual diagram showing an example of a work management table TB. The work management table TB is a table for managing the work performed by workers. The work management table TB stores, for example, the work name, machine name, work date and time, worker ID, standard procedure ID, supplementary information, equipment data, and destination for sending the procedure, all in association with each other. The work name is the name of the work that the worker is to perform. The machine name is the name of the machine being worked on. The work date and time includes the start date and time of the work. The worker ID is an identifier for identifying the worker. Multiple worker IDs may be assigned to a single work. The standard procedure ID is an identifier for identifying the standard procedure for that work. Standard procedure documents are created separately by the work manager, etc., and registered in the standard procedure database DB2 in association with the standard procedure ID.

[0055] Supplementary information is supplementary information related to the work. Supplementary information may include, for example, service guides, instruction manuals, installation manuals, past failure cases, and recent trend information regarding the equipment being worked on. Supplementary information may also include information such as the equipment's configuration, intended use, years of service, and past repair history. Supplementary information may also include information about peripheral equipment connected to the equipment being worked on (fan coil units, thermostats, filters, third-party components and equipment that are not our own products, etc.). Supplementary information may also include information about the property where the equipment being worked on is installed (purpose of use of the property, information about the surrounding environment such as room temperature, wind speed, and solar radiation conditions), information about the work site (installation height, distance from walls, and other installation location information), and information about time constraints in the work. Supplementary information may be registered as links to electronic documents, or as text data.

[0056] Equipment data is data obtained about the equipment being worked on, and includes data such as control values ​​input to the equipment and measured values ​​measured for the equipment. Equipment data may also include image data captured in advance at the work site and sound data recorded in advance at the work site. Equipment data is registered as a link to electronic data.

[0057] The destination for the procedure manual indicates the destination for the revised procedure manual. The server device 20 uses the language model MD to generate a revised procedure manual according to the worker's attributes and sends the generated revised procedure manual to the wearable device 10 worn by that worker. The communication address of the wearable device 10 is specified as the destination for the revised procedure manual.

[0058] Of the information registered in the work management table TB, the work name, model name, work date and time, worker ID, standard procedure ID, and procedure recipient are registered in advance by the work manager or other relevant personnel. Supplementary information is not mandatory, but if there is any information that should be registered, it will be registered in advance by the work manager or other relevant personnel. Equipment data is also not mandatory, but if it is obtained at the work site before the start of work, it will be registered in advance by the worker or other relevant personnel.

[0059] Figure 5 is a conceptual diagram showing an example of an attribute database DB1. The attribute database DB1 is a database that stores attribute information of workers in association with worker IDs. In the example in Figure 5, the worker's skill level, qualifications, work speed, and evaluation of the worker (frequent service rate, free multiple parts replacement rate, paid service rate, average number of visits, and call center (CC) inquiry rate) are stored as attribute information. The attribute database DB1 only needs to store any one of these attribute information.

[0060] Skill level is an indicator of how proficient a worker is in the overall work. In the example in Figure 5, two levels, high and low, are used as indicators, but three or more levels may be used, or years of experience may be used as an indicator. Furthermore, there does not need to be one indicator for each worker; for example, skill levels may be set for each type of product, such as household air conditioners, commercial air conditioners, and air purifiers, or for each type of refrigeration system, such as direct expansion type and indirect expansion type. Qualifications indicate whether or not a worker has a public qualification or a qualification obtained through an in-house certification exam. It may also include information on the types of qualifications the worker holds, not just whether or not they have a qualification. Work speed represents the number of days it takes to complete the work. Alternatively, relative indicators such as fast / slow may be used as work speed.

[0061] The multiple service rate represents the rate at which a customer requests the same service again after the initial work has been completed. The free multiple parts replacement rate represents the rate at which parts were replaced without charging the customer. This metric is based on the principle that if part A is faulty but part B is replaced and the problem is not resolved, the cost of part B cannot be charged. The paid service rate represents the rate at which work that is not free of charge is performed and for which a fee can be charged is performed. The average number of visits represents the average number of visits to the work site per job. The call center inquiry rate represents the rate at which the call center is contacted again after the work has been completed.

[0062] In addition to the information shown in Figure 5, worker attribute information may also include subjective customer evaluations (such as friendliness), self-reported evaluations, number of jobs performed in the past, efforts to make the shortest possible visit, efforts to adjust remaining work as quickly as possible, efforts to reduce the number of estimates completed, efforts to complete estimate-ordered work as quickly as possible, proposals to customers (such as proposals for refurbished substrates and use of refrigerant leak prevention agents), promotion of the use of specific parts (right-exit heat exchangers), and the rate of parts returned.

[0063] Figure 6 is a conceptual diagram showing an example of the standard procedure database DB2. The standard procedure database DB2 is a database that stores standard procedure manuals associated with standard procedure manual IDs. Standard procedure manuals exist for each type and classification of equipment to be worked on, and for each type of work (inspection work, repair work, installation work, commissioning work, disposal work, etc.), and consist of standard work procedures for that work. The standard procedure database stores, for example, the procedure number, type, category, and body text associated with each other. The procedure number represents the order of the procedure. The type represents the distinction between guidance and questions. If the type is a question, it indicates a procedure that requires a response from the worker. In Figure 6, the questions exemplified by procedure numbers "02" to "06" all indicate that they can be answered with YES or NO. The category represents the object to be worked on (outdoor unit, indoor unit, controller, tank, etc.). The body text is a text representation of the voice guidance output from the wearable device 10, and describes the work that the worker should perform in that procedure.

[0064] In Figure 6, examples of questions that can be answered with YES or NO are shown. However, questions that require numerical input such as temperature or pressure, or questions that require text input such as error codes or free-form descriptions, may also be registered.

[0065] The attribute database DB1 and the standard procedure database DB2 may be located outside the server device 20, provided they are accessible from the server device 20. Alternatively, the attribute database DB1 and the standard procedure database DB2 may be accessible from the LLM server 30.

[0066] The server device 20 generates revised procedure manuals using a language model MD that references worker attribute information and standard procedure manuals for the work performed by the worker. Figure 7 is an explanatory diagram illustrating an example of revised procedure manual generation. The worker attribute information referenced by the language model MD and the standard procedure manuals for the work performed by the worker are specified by the server device 20. For example, when generating a revised procedure manual for a task (inspection work) registered as the first record in the work management table TB in Figure 4, the server device 20 searches the attribute database DB1 using the worker ID (USER001 in this example) as the search key and obtains the attribute information of the corresponding worker. The server device 20 specifies the attribute information obtained from the attribute database DB1 as the worker attribute information to be used when generating the revised procedure manual. Also, by referring to the work management table TB, it can be seen that the standard procedure manual ID for the above inspection work is "SM001", so the server device 20 specifies this "SM001" standard procedure manual as the standard procedure manual to be used when generating the revised procedure manual.

[0067] The language model MD generates revised procedure manuals by referring to the worker attribute information and standard procedure manuals specified by the server device 20. The language model MD is a deep learning model configured to output answer sentences (text data) in response to the input of question sentences (text data). Existing language models (large-scale language models) such as GPT-4 (Generative Pretrained Transformer 4), LLaMA (Large Language Model Meta AI), and BERT (Bidirectional Encoder Representations from Transformers) can be used for the language model MD. The language model MD is not limited to the above language models, but may also be a unique language model that has been tuned from an existing language model. Furthermore, the language model MD may be a multimodal language model that can accept inputs such as measurement data, imaging data, and acoustic data obtained regarding the equipment being worked on, in addition to text data.

[0068] The revised procedure manuals generated by the language model MD are revisions of the standard procedure manuals according to the worker's attribute information. A worker with worker ID "USER001" has attributes such as high skill level, qualifications, fast work speed, and high work performance evaluation. Therefore, as shown in Figure 7, the explanation of some work procedures (e.g., procedure 01) may be omitted, or some work procedures (e.g., procedure 03) may be skipped. Even in revised procedure manuals for highly skilled workers, additional procedures (procedure 04 + α) may be added from the standard procedure manual. Although not shown in Figure 7, for workers with attributes such as low skill level, unqualified, slow work speed, and low work performance evaluation, the explanation of some work procedures may be made more detailed, or preliminary work procedures may be added for some work procedures.

[0069] Figure 7 illustrates an example of revising a standard procedure manual by referring to worker attribute information. However, it is also possible to specify that the system should refer to supplementary information or equipment data registered in the work management table TB in addition to worker attribute information, and generate a revised procedure manual that takes the supplementary information or equipment data into account.

[0070] For example, if supplementary information such as past failure cases or recent trend information is specified, additional procedures may be added to prevent common malfunctions in the equipment being worked on. Similarly, if information about peripheral equipment, the property, or the work site is specified, additional procedures based on information that can only be determined at the work site may be added. Furthermore, if equipment data is specified, additional procedures may be added to resolve existing malfunctions in the equipment being worked on.

[0071] The server device 20 generates input data (prompts) for the language model MD in order to generate the revised procedure manual as described above. Figure 8 is a schematic diagram showing an example of a prompt to be input to the language model MD. As prompts to be input to the language model MD, the server device 20 generates prompts that, for example, instruct that the role is that of a work process manager, that the worker attribute information listed in the reference information column should be referred to, that the base standard procedure manual (in the example in Figure 8, the standard procedure manual for SM0001) is registered in the standard procedure database DB2, and that a revised procedure manual should be created by revising the standard procedure manual according to the worker's attributes. The server device 20 generates such prompts and sends them to the LLM server 30 via the communication network NW.

[0072] The LLM server 30 inputs the prompt received from the server device 20 into the language model MD and obtains the revised procedure document output as a response from the language model MD. The LLM server 30 returns the revised procedure document to the server device 20. The server device 20 receives the revised procedure document sent from the LLM server 30. The server device 20 can generate the revised procedure document by utilizing the language model MD provided by the LLM server 30.

[0073] Figure 8 shows an example of a prompt for creating a revised procedure manual by referring to worker attribute information. The same applies to prompts for creating a revised procedure manual by referring to supplementary information or equipment data in addition to attribute information. In other words, the supplementary information or equipment data to be referred to should be listed in the reference information column. If the supplementary information or equipment data to be referred to exists as a data file, a link to that data file may be specified as the reference information.

[0074] The following describes the processes performed by the server device 20. Figure 9 is a flowchart illustrating the procedure for processing performed by the server device 20. The control unit 21 of the server device 20 refers to the work management table TB and selects the work for which a revised procedure manual should be created (step S101). For example, the control unit 21 checks the work date and time column of the work management table TB and selects the work that will start in the near future.

[0075] The control unit 21 reads the worker ID and standard procedure ID registered in association with the selected work from the work management table TB (step S102). The control unit 21 uses the read worker ID as a search key to search the attribute database DB1 and reads the attribute information of the corresponding worker from the attribute database DB1 (step S103).

[0076] The control unit 21 generates prompts to be input to the language model MD based on the information read in steps S102 and S103 (step S104). As prompts to be input to the language model MD, the control unit 21 generates prompts that, for example, instruct that the role is that of a work process manager, that the worker's attribute information listed in the prompt should be referenced, that the base standard procedure manual is registered in the standard procedure database DB2, and that a revised procedure manual should be created by revising the standard procedure manual according to the worker's attributes. If supplementary information or equipment data is registered in the work management table TB, the control unit 21 may generate prompts that specify the reference to supplementary information or equipment data in addition to attribute information.

[0077] The control unit 21 sends the generated prompt to the LLM server 30 via the communication network NW (step S105). When the LLM server 30 receives the prompt, it inputs the received prompt into the language model MD and generates a response statement (revised procedure document in this embodiment) using the language model MD. The LLM server 30 returns the revised procedure document obtained from the language model MD to the server device 20 via the communication network NW.

[0078] The control unit 21 receives the revised procedure document returned from the LLM server 30 after sending a prompt (step S106). The control unit 21 identifies the destination of the received revised procedure document by referring to the work management table TB (step S107). The control unit 21 sends the revised procedure document received in step S106 to the destination (wearable device 10) identified in step S107 (step S108).

[0079] The wearable device 10 receives a revised procedure manual transmitted from the server device 20 and provides voice guidance according to the received revised procedure manual.

[0080] As described above, in this embodiment, revised procedures for each worker can be automatically created based on the standard procedure manual, thus reducing the man-hours required to create the procedure manual. For highly skilled workers, unnecessary voice guidance and work procedures can be omitted. For inexperienced workers, detailed explanations of the work procedures can be used to improve the quality and speed of the work.

[0081] In this embodiment, a single language model MD is used to generate revised procedure manuals. However, for example, a multi-agent mechanism may be used in which a first language model configured to output procedure information based on worker attribute information, a second language model configured to output procedure information based on supplementary information, a third language model configured to output procedure information based on equipment data, and a main language model that generates a final revised procedure manual to be presented to the worker based on the procedure information output from the first to third language models, with these language models collaborating to generate the revised procedure manual. The first language model generates procedures with added (or omitted) steps or explanations depending on the worker's attributes. The second language model may generate additional procedures to prevent malfunctions that are likely to occur in the target equipment, based on past cases such as failures and recent trend information. The third language model may generate additional procedures to resolve malfunctions that have already occurred in the equipment. The main language model can generate a single revised procedure manual by integrating these procedures.

[0082] In this embodiment, the system is configured to generate revised procedure manuals according to the worker's attribute information. However, the server device 20 may also accept the worker's own requests, generate a prompt instructing the server to create a revised procedure manual that incorporates the worker's requests, and send the generated prompt to the LLM server 30. The worker can use their terminal to give the server device 20 requests, for example, to include more explanations or fewer explanations. The server device 20 can generate a revised procedure manual with more (or less) explanations by sending the prompt generated according to such requests to the LLM server 30.

[0083] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims, not in the sense described above, and all modifications within the sense and scope equivalent to the claims are intended. [Explanation of Symbols]

[0084] 10 Wearable Devices 20 Server Devices 21 Control Unit 22 Memory section 23 Communications Department 24 Control section 25 Display section 30 LLM Servers PG Procedure Generation Program TB Work Management Table DB1 Attribute Database DB2 Standard Procedure Database MD Language Model

Claims

1. Obtain attribute information indicating the attributes of the worker performing the task, Using one or more language models that refer to acquired attribute information and standard procedure information indicating the standard procedure for the said work, work procedure information corresponding to the worker's attributes is generated. Output the generated work procedure information. An information processing method in which processing is performed by a computer.

2. The attribute information includes at least one of the worker's skill level, qualifications, work speed, and evaluation of the worker. The information processing method according to claim 1.

3. The aforementioned evaluation includes at least one of the following: the rate of frequent service visits, the rate of free replacement of multiple parts, the rate of paid service, the average number of visits, and the call center inquiry rate. The information processing method according to claim 2.

4. Further obtain at least one of the following: information regarding the aforementioned work, information regarding the equipment to be worked on, information regarding past cases in the aforementioned work, and supplementary information regarding the aforementioned standard procedure. Using the attribute information, the standard procedure information, and a language model that references at least one acquired piece of information, work procedure information corresponding to the worker's attributes is generated. The information processing method according to claim 1, wherein the processing is performed by the computer.

5. The information relating to the said work includes at least one of the work site information and information regarding the time constraints in said work. The information processing method according to claim 4.

6. The information relating to the aforementioned device includes at least one of the device's configuration information, measured values ​​measured for the device, control values ​​input to the device, and the intended use of the device. The information processing method according to claim 4.

7. Acquire at least one of the image data captured at the work site of the aforementioned work and the sound data recorded at the work site. Using the attribute information, the standard procedure information, and a language model that references at least one of the acquired data, work procedure information corresponding to the worker's attributes is generated. The information processing method according to claim 1, wherein the processing is performed by the computer.

8. We accept worker requests for work procedure information. The language model is used to generate work procedure information in response to the worker's request. The information processing method according to claim 1, wherein the processing is performed by the computer.

9. It comprises at least one control unit, The control unit, Obtain attribute information indicating the attributes of the worker performing the task, Using one or more language models that refer to acquired attribute information and standard procedure information indicating the standard procedure for the said work, work procedure information corresponding to the worker's attributes is generated. Output the generated work procedure information. Information processing device.

10. The worker's terminal device, A server device that is communicatively connected to the aforementioned terminal device and has a storage unit that stores standard procedure information indicating the standard procedure for the work, Equipped with, The server device is Obtain attribute information indicating the attributes of the worker who performed the aforementioned work, Using one or more language models that refer to the acquired attribute information and the standard procedure information stored in the storage unit, work procedure information corresponding to the attributes of the worker performing the work is generated. Output the generated work procedure information. Information processing system.

11. Obtain attribute information indicating the attributes of the worker performing the task, Using one or more language models that refer to acquired attribute information and standard procedure information indicating the standard procedure for the said work, work procedure information corresponding to the worker's attributes is generated. Output the generated work procedure information. A computer program that causes a computer to perform a process.

Citation Information

Patent Citations

  • Information processing apparatus, information processing system, information processing method, and information processing program

    JP2017126292A

  • Manual management device, manual management system, and control program

    JP2020004274A

  • Maintenance work support device, maintenance work support method, maintenance work support program, and recording medium

    JP2020035497A

  • Information processing method, computer program, and information processing device

    WO2023182424A1

  • Inspection work support system, method, and program

    JP2019057099A