Preventive maintenance system

Abstract

A preventive maintenance system (1A) comprises: an information collection device (30) that collects device information, which is information on the state of an air conditioner (20); and a complaint response notification device (10A) that, on the basis of the device information, transmits, to a user terminal (40), an FAQ (16A) in which a complaint not pertaining to a failure of the air conditioner (20) is associated with an answer to the complaint. The complaint response notification device (10A) has: an estimation unit (12) that estimates a complaint on the basis of the device information and reads out an FAQ (16A) corresponding to the estimated complaint from among the plurality of FAQs (16A); and a communication unit (13) that transmits the read-out FAQ (16A) to the user terminal.

Description

Preventive Maintenance System

[0001] The present disclosure relates to a preventative maintenance system that reduces maintenance services for equipment.

[0002] Equipment such as air conditioners may malfunction during use, and if requested by a user, a maintenance service technician will check for the malfunction. Such maintenance service by a maintenance service technician takes time and effort, so there is a demand for more efficient maintenance services.

[0003] The automatic response server device described in Patent Document 1 generates a response prompting the user to input the water heater error code when the voice content from the telephone indicates that an error code is displayed on the user's water heater. If the error code input in response to the generated response is an error code that requires water heater repair, the automatic response server device generates a response prompting the user to input a decision on whether or not to carry out repairs.

[0004] Japanese Patent Application Laid-Open No. 2021-103843

[0005] However, with the technology of Patent Document 1, calls are accepted from users without the details of the malfunction being understood, so all calls must be answered, and the number of cases for which maintenance services for equipment are handled cannot be reduced.

[0006] The present disclosure has been made in consideration of the above, and aims to provide a preventive maintenance system that can reduce the number of maintenance service cases for equipment.

[0007] In order to solve the above-mentioned problems and achieve the objectives, the preventive maintenance system of the present disclosure includes an information collection device that collects equipment information, which is information on the status of equipment, and a complaint response notification device that transmits maintenance response information, in which complaints unrelated to equipment failures are associated with responses to the complaints, to a user terminal based on the equipment information. The complaint response notification device has an estimation unit that estimates complaints based on the equipment information and reads out maintenance response information corresponding to the estimated complaint from multiple pieces of maintenance response information, and a communication unit that transmits the read out maintenance response information to the user terminal.

[0008] The preventive maintenance system according to the present disclosure has the effect of reducing the number of maintenance service requests for equipment.

[0009] FIG. 1 is a diagram showing the configuration of a preventive maintenance system according to a first embodiment; FIG. 2 is a diagram for explaining the configuration of FAQs (Frequently Asked Questions) stored by the preventive maintenance system according to the first embodiment; FIG. 3 is a flowchart showing the processing procedure of processing executed by the preventive maintenance system according to the first embodiment; FIG. 4 is a diagram showing the configuration of a preventive maintenance system according to a second embodiment; FIG. 5 is a diagram for explaining an example of an FAQ generation prompt used by the preventive maintenance system according to the second embodiment; FIG. 6 is a diagram for explaining an example of maintenance data used by the preventive maintenance system according to the second embodiment;

[0010] A preventive maintenance system according to an embodiment of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0011] First Embodiment. Fig. 1 is a diagram showing the configuration of a preventive maintenance system according to a first embodiment. The preventive maintenance system 1A is a system that reduces maintenance services for equipment used by users. The preventive maintenance system 1A reduces the number of visits by maintenance service personnel when such visits are unnecessary, such as when the user has a misunderstanding, thereby realizing labor-saving in maintenance services in the after-sales service industry. Furthermore, the preventive maintenance system 1A improves user satisfaction by presenting simple maintenance methods to users in an easy-to-understand manner.

[0012] In the following, we will explain the case where the target equipment (equipment used by the user) for which the preventive maintenance system 1A will reduce maintenance services is air conditioning equipment 20, but the target equipment for which maintenance services will be reduced may also be equipment other than air conditioning equipment 20 (water heaters, refrigerators, washing machines, etc.).

[0013] The preventive maintenance system 1A includes a complaint handling notification device (claim handling notification device) 10A, air conditioning equipment 20, an information collection device 30, a user terminal 40, an FAQ DB (Database) 2, and an equipment information DB 8.

[0014] The complaint handling notification device 10A, the information collection device 30, and the user terminal 40 are connected via a network 9 such as the Internet, and transmit and receive information via the network 9. Note that the air conditioning equipment 20 may also be connected to the network 9. In the following, the fact that the network 9 is involved in the transmission and reception of information within the preventive maintenance system 1A may be omitted.

[0015] The FAQDB2, which is a maintenance response database, is connected to the complaint response notification device 10A. The FAQDB2 stores a plurality of FAQs 16A, which are FAQ data. The FAQs 16A are information that associates complaints received from users with responses to those complaints (such as how to deal with the complaints). The complaints in the FAQs 16A include complaints that are frequently received from users and complaints that are expected to be received from users. The complaints in the FAQs 16A may also include questions. The FAQs 16A include complaints that are not related to failures of the air conditioning equipment 20. In other words, the FAQs 16A include complaints and responses that are not related to failures and do not require a response from a maintenance service technician.

[0016] FAQ 16A is configured to include, for example, a category, symptoms, causes, and countermeasures. A complaint in FAQ 16A includes a category and symptoms, and an answer to FAQ 16A (hereinafter sometimes referred to as maintenance response information) includes a cause and countermeasures. The categories, symptoms, causes, countermeasures, and the like will be described later.

[0017] The complaints included in the FAQ 16A are, for example, malfunctions (categories, symptoms) of the air conditioning equipment 20, and the answers corresponding to these complaints are, for example, methods of dealing with the malfunctions (causes, measures) of the air conditioning equipment 20. The FAQ DB 2 stores FAQs 16A for various types of air conditioning equipment 20.

[0018] The equipment information DB 8 is connected to the information collecting device 30. The equipment information DB 8 stores equipment information relating to the status of the air conditioning equipment 20 and the like collected by the information collecting device 30. The equipment information includes the operating status of the air conditioning equipment 20, the operating history of the air conditioning equipment 20, information on the environment in which the air conditioning equipment 20 is placed, information on the operation history of the air conditioning equipment 20, and the like. The equipment information DB 8 stores equipment information for various types of air conditioning equipment 20.

[0019] The information collection device 30 is connected to the air conditioning equipment 20. The information collection device 30 is a computer that collects equipment information from the air conditioning equipment 20 and provides it to the complaint response notification device 10A. The information collection device 30 stores the collected equipment information in the equipment information DB 8.

[0020] The information collection device 30 has an information acquisition unit 31, an equipment information storage unit 32, and a communication unit 33. The information acquisition unit 31 acquires, from the air conditioning equipment 20, equipment information about the equipment being managed.

[0021] The device information storage unit 32 is a memory or the like that stores the device information acquired by the information acquisition unit 31. The communication unit 33 transmits the device information acquired by the information acquisition unit 31 and stored in the device information storage unit 32 to the complaint handling notification device 10A.

[0022] The complaint handling notification device 10A is a computer that estimates complaints that may be made by users of the air conditioning equipment 20, generates responses corresponding to the estimated complaints, and transmits the responses to the user terminal 40.

[0023] The complaint handling notification device 10A has an FAQ storage unit 11, an estimation unit 12, and a communication unit 13. The FAQ storage unit 11 stores, among the FAQs 16A in the FAQ DB 2, the FAQs 16A of the air conditioning equipment 20 that is the subject of management.

[0024] The estimation unit 12 detects an abnormal state of the air conditioning equipment 20 based on the equipment information sent by the information collection device 30 to the complaint response notification device 10A, and estimates a complaint that may be made by the user based on the detection result. That is, the estimation unit 12 estimates a question corresponding to the equipment information. An abnormal state of the air conditioning equipment 20 may be a state in which the user is using the air conditioning equipment 20 in an incorrect manner, a state in which the user is using the air conditioning equipment 20 with a mistaken understanding, etc.

[0025] If the equipment information includes information about a filter error, the estimation unit 12 detects that the filter is dirty as an abnormal state of the air conditioning equipment 20. Then, the estimation unit 12 estimates that the dirty filter may prevent the indoor temperature from decreasing, leading to complaints from users that the room is not "cool."

[0026] Furthermore, the estimation unit 12 may detect, for example, a temperature difference between the set temperature of the air conditioning equipment 20 and the actual room temperature based on the equipment information as an abnormal state of the air conditioning equipment 20. In this case, if the temperature difference is equal to or greater than a reference value and the equipment information does not include information about a filter error, the estimation unit 12 estimates that the indoor temperature will not drop due to the door being opened or closed, and that there is a possibility that a user will complain that the room is not cold.

[0027] The estimation unit 12 reads out an FAQ 16A corresponding to the estimated complaint from the FAQ storage unit 11. For example, the estimation unit 12 selects and reads out information including a complaint that "it's not cooling" and a response (maintenance response) to this complaint from the FAQ storage unit 11 as the FAQ 16A corresponding to the air conditioning equipment 20. In other words, the estimation unit 12 reads out, from the FAQ storage unit 11, maintenance response information corresponding to the air conditioning equipment 20 from a plurality of pieces of maintenance response information in which complaints and responses are associated with each other.

[0028] The communication unit 13 receives the device information transmitted by the information collection device 30 via the network 9. The communication unit 13 also transmits the maintenance response information read by the estimation unit 12 to the user terminal 40 via the network 9.

[0029] In this way, the complaint response notification device 10A estimates complaints that may be made by users of the air conditioning equipment 20 from the equipment information, obtains maintenance response information corresponding to the estimated complaints from the FAQ 16A, and transmits it to the user terminal 40.

[0030] The user terminal 40 is a terminal used by a user of the air conditioning equipment 20. The user terminal 40 is, for example, a smartphone, a tablet terminal, or a mobile phone. The user terminal 40 receives and displays maintenance response information. This allows the user of the air conditioning equipment 20 to check maintenance response information that addresses complaints that are not related to a malfunction of the air conditioning equipment 20.

[0031] Here, a description will be given of the FAQ 16A stored in the FAQ storage unit 11. Fig. 2 is a diagram for explaining the configuration of the FAQ stored in the preventive maintenance system according to the first embodiment.

[0032] The FAQ storage unit 11 stores, for example, an FAQ 16A having a structure as shown in Fig. 2. The FAQ 16A is a graph database made up of categories, symptoms, causes, countermeasures, etc. A graph database is a database with a graph structure, and the data structure is network-like rather than relational.

[0033] In the FAQ 16A, categories are associated with symptoms, symptoms are associated with causes, and causes are associated with countermeasures. The categories indicate the classification of the content of the FAQ 16A. Examples of categories include "user maintenance" for users to perform maintenance, and "serviceman maintenance" for maintenance by service personnel. In the first embodiment, the FAQ 16A includes "user maintenance."

[0034] The symptom is a symptom of the air conditioning equipment 20 that corresponds to the complaint. Examples of the symptom include "not cooling" and "not heating." The symptom of "not cooling" indicates that the room is not cooled by the operation of the air conditioning equipment 20, and the symptom of "not heating" indicates that the room is not heated by the operation of the air conditioning equipment 20.

[0035] The cause is the cause of the complaint. Examples of causes include "performance degradation due to a clogged filter" and "exceeding the operating capacity or room temperature change due to door opening and closing." For example, if the symptom is "not cooling," the causes are "performance degradation due to a clogged filter" and "exceeding the operating capacity or room temperature change due to door opening and closing."

[0036] Countermeasures are measures (actions) taken in response to complaints. Examples of countermeasures are "cleaning the filter" and "closing the door or window." For example, if the cause is "performance degradation due to a clogged filter," the countermeasure is "cleaning the filter." If the cause is "exceeding the operating capacity, causing room temperature changes due to door opening and closing," the countermeasure is "close the door or window."

[0037] For example, when the estimation unit 12 estimates a complaint such as "it's not cooling" based on the equipment information, it reads out answers to this complaint, such as "clean the filter" or "close the door or window," from the FAQ memory unit 11.

[0038] Next, the processing procedure of the processing executed in the preventive maintenance system 1A will be described. Fig. 3 is a flowchart showing the processing procedure of the processing executed in the preventive maintenance system according to the first embodiment. Here, as the operation of the preventive maintenance system 1A, the operation of the complaint response notification device 10A and the information collection device 30 will be described.

[0039] The complaint handling notification device 10A acquires the FAQ 16A of the air conditioning equipment 20 from the FAQ DB 2 (step S10). The complaint handling notification device 10A stores the FAQ 16A in the FAQ storage unit 11.

[0040] The information acquisition unit 31 of the information collection device 30 acquires equipment information including operation information of the air conditioning equipment 20 from the air conditioning equipment 20 (step S20). Note that either the process of step S10 or the process of step S20 may be executed first. The communication unit 33 of the information collection device 30 transmits the equipment information to the complaint response notification device 10A.

[0041] The estimation unit 12 of the complaint handling notification device 10A determines whether or not an abnormal state has been detected based on the device information (Step S30). If the estimation unit 12 does not detect an abnormal state (No in Step S30), the estimation unit 12 continues the process of determining whether or not an abnormal state has been detected (Step S30).

[0042] When the estimation unit 12 detects an abnormal state (Yes in step S30), the estimation unit 12 estimates a complaint (content of a claim) that may be made by the user based on the abnormal state (step S40). The estimation unit 12 reads the estimated complaint and a response to the complaint from the FAQ storage unit 11 as maintenance response information.

[0043] The communication unit 13 notifies the user of the read maintenance response information (step S50). That is, the communication unit 13 transmits the read maintenance response information to the user terminal 40 via the network 9.

[0044] For example, the information acquisition unit 31 acquires, as device information, information on the power supply status, information on the operation mode, information on the set temperature, and information on the room temperature. For example, when the power supply status information indicates that the power is on, the operation mode information indicates that the air conditioner is in cooling mode, and the temperature difference between the set temperature and the room temperature remains unchanged for one hour or more, the estimation unit 12 detects an abnormal state in which the room temperature does not change, and estimates a complaint that the room temperature is not being cooled properly based on this abnormal state. That is, the estimation unit 12 estimates and detects a complaint indicating that the room temperature is not being cooled properly as a complaint that may be made by a user.

[0045] For example, the estimation unit 12 acquires from the FAQ 16A answers associated with a complaint indicating that the indoor temperature is difficult to cool in the FAQ 16A. If the answers associated with the complaint indicating that the indoor temperature is difficult to cool include "Please check that the door is not left open" and "Please change the air speed to "automatic" or "strong", the estimation unit 12 acquires these answers from the FAQ 16A and sends them to the communication unit 13.

[0046] The communication unit 13 transmits the response to the user terminal 40. As a result, the complaint response notification device 10A notifies the user of the expected complaint and the response to the complaint as maintenance response information via an app (application). The app notification (push notification) is a function that notifies the user of the maintenance response information. The user terminal 40 displays the maintenance response information via the app notification.

[0047] In this way, the preventive maintenance system 1A estimates the complaint corresponding to the equipment information and transmits the estimated complaint and the answer to the FAQ 16A corresponding to the complaint as maintenance response information to the user terminal 40. This allows the preventive maintenance system 1A to notify the user of the answer corresponding to the estimated complaint when there is a misunderstanding by the user rather than a product abnormality in the air conditioning equipment 20, or when there is a maintenance response that the user can easily perform. As a result, the preventive maintenance system 1A can reduce requests for maintenance service due to user misunderstanding, etc., and therefore the number of maintenance service responses can be reduced.

[0048] For example, a user may mistakenly believe that the problem is due to a malfunction of the air conditioning equipment 20, even though the room is not cooled due to the door being opened and closed more frequently than the product's operating capacity. Maintenance that can be easily performed by the user includes opening and closing the door, adjusting the airflow volume, etc.

[0049] The preventive maintenance system 1A notifies the user of maintenance response information before a complaint is received, for example, if the user has a misunderstanding or if there is a maintenance response that the user can easily perform. As a result, the preventive maintenance system 1A improves user satisfaction by providing explanations to the user, and reduces maintenance costs and maintenance services by reducing unnecessary visits by maintenance service personnel. Furthermore, if the user can easily perform maintenance, on-site work by a maintenance service personnel is unnecessary, and the preventive maintenance system 1A can reduce and streamline maintenance services.

[0050] The preventive maintenance system 1A can also be applied to other maintenance services such as a maintenance service for diagnosing the refrigerant circuit of the air conditioning equipment 20 and a maintenance service for detecting faulty parts.

[0051] As described above, according to the first embodiment, the complaint response notification device 10A estimates the complaint based on the device information, and transmits maintenance response information corresponding to the estimated complaint to the user terminal 40. As a result, the preventive maintenance system 1A can prevent unnecessary business trips by maintenance service personnel, thereby reducing the number of maintenance service cases.

[0052] Second Embodiment Next, a second embodiment will be described with reference to Figures 4 to 6. In the second embodiment, the FAQ 16A is automatically generated using a language model such as a large-scale language model.

[0053] Fig. 4 is a diagram showing the configuration of a preventive maintenance system according to embodiment 2. Of the components in Fig. 4, those that achieve the same functions as those in the preventive maintenance system 1A according to embodiment 1 shown in Fig. 1 are given the same reference numerals, and duplicated explanations will be omitted.

[0054] The preventive maintenance system 1B of the second embodiment is a system that reduces maintenance services for equipment used by users, similar to the preventive maintenance system 1A of the first embodiment. The preventive maintenance system 1B includes a complaint response notification device 10A, an air conditioning device 20, an information collection device 30, a user terminal 40, an FAQ DB 2, an FAQ generation device 50, an FAQ generation prompt DB 3, a large-scale language model DB 4, and a maintenance data DB 5.

[0055] An FAQ generation device 50, which is a maintenance information generation device, is connected to a network 9 and transmits and receives information via the network 9. The FAQ generation device 50 is also connected to an FAQ generation prompt DB 3, a large-scale language model DB 4, and a maintenance data DB 5.

[0056] The maintenance data DB 5 stores maintenance data (maintenance data 55, which will be described later). The maintenance data 55 is data indicating the work performed by a maintenance worker in response to an abnormal state of the air conditioning equipment 20.

[0057] The large-scale language model DB4 stores a large-scale language model (first large-scale language model). The large-scale language model of the second embodiment is a natural language processing model trained using a large amount of text data. Examples of large-scale language models include ChatGPT by OpenAI, Inc. and Bard (registered trademark) by Google (registered trademark).

[0058] The FAQ generation prompt DB 3 stores FAQ generation prompts, which are commands (instructions, questions, etc.) input by a user in an interactive system such as a dialogue with AI (Artificial Intelligence) or a command line interface (CLI).

[0059] In the first embodiment, the FAQ 16A is manually generated based on information contained in an existing manual, etc. In the second embodiment, the FAQ generation device 50 automatically generates the FAQ 16A using the maintenance data 55, the large-scale language model, and the FAQ generation prompt.

[0060] At least one of the FAQ generation prompt DB 3, the large-scale language model DB 4, and the maintenance data DB 5 may be located within the complaint handling notification device 10A.

[0061] The FAQ generation device 50 includes a maintenance response estimation unit 51, a maintenance data acquisition unit 52, and a communication unit 53. The maintenance data acquisition unit 52 acquires maintenance data 55 from the maintenance data DB 5.

[0062] The maintenance response estimation unit 51 reads out an FAQ generation prompt from the FAQ generation prompt DB 3. The maintenance response estimation unit 51 also reads out a large-scale language model from the large-scale language model DB 4. The maintenance response estimation unit 51 generates an FAQ 16A from the maintenance data 55 by causing the large-scale language model to execute the FAQ generation prompt. That is, the maintenance response estimation unit 51 uses the large-scale language model to generate an FAQ 16A corresponding to the maintenance data 55 from the maintenance data 55.

[0063] The communication unit 53 transmits the FAQ 16A to the complaint response notification device 10A via the network 9. The communication unit 53 may transmit the FAQ 16A to the FAQ DB 2 via the complaint response notification device 10A, or may transmit the FAQ 16A directly to the FAQ DB 2.

[0064] An example of an FAQ generation prompt will now be described. Fig. 5 is a diagram for explaining an example of an FAQ generation prompt used by the preventive maintenance system according to the second embodiment. The FAQ generation prompt 54 is written in the following format, for example: #Request #Role #Format #Rule

[0065] The "#request" field contains the details of the job request, and the "#role" field contains the role that the large-scale language model will play. The "#format" field contains the format of the FAQ 16A to be generated, and the "#rule" field contains the rules for generating the FAQ 16A.

[0066] The complaint handling notification device 10A stores the FAQ 16A generated from the maintenance data 55 in the FAQ storage unit 11. Fig. 6 is a diagram for explaining an example of maintenance data used by the preventive maintenance system according to the second embodiment.

[0067] The maintenance data 55 associates abnormal states with measures to be taken. The abnormal states are information indicating the state in which the air conditioning equipment 20 is being used, and the measures are information on the work performed by the maintenance worker.

[0068] The maintenance response estimation unit 51 causes a large-scale language model such as ChatGPT by OpenAI or Bard (registered trademark) by Google (registered trademark) to execute a command statement such as the FAQ generation prompt 54, and generates the FAQ 16A from the maintenance data 55. Note that the maintenance response estimation unit 51 and the maintenance data acquisition unit 52 may be separate devices.

[0069] In this way, the preventive maintenance system 1B of embodiment 2 automatically generates FAQ 16A from maintenance data 55, thereby reducing the labor costs of generating FAQ 16A and providing users with the latest maintenance response information.

[0070] Third Embodiment Next, a third embodiment will be described with reference to Figures 7 and 8. In the third embodiment, at least one of an image and a video is associated with the FAQ 16A.

[0071] Fig. 7 is a diagram showing the configuration of a preventive maintenance system according to embodiment 3. Of the components in Fig. 7, those that achieve the same functions as those in the preventive maintenance system 1A according to embodiment 1 shown in Fig. 1 are given the same reference numerals, and duplicated explanations will be omitted.

[0072] Similar to the preventive maintenance system 1A of the first embodiment, the preventive maintenance system 1C of the third embodiment is a system that reduces maintenance services for equipment used by users. The preventive maintenance system 1C includes a complaint response notification device 10C, air conditioning equipment 20, an information collection device 30, a user terminal 40, an FAQ DB2, and an image and video DB6. As such, the preventive maintenance system 1C includes the image and video DB6 in addition to the components included in the preventive maintenance system 1A. Furthermore, compared to the preventive maintenance system 1A, the preventive maintenance system 1C includes a complaint response notification device 10C instead of the complaint response notification device 10A. Note that the image and video DB6 may be located within the complaint response notification device 10C.

[0073] The complaint handling notification device 10C is connected to the image and video DB 6. The image and video DB 6 stores at least one of images (image data) and videos (video data) corresponding to the maintenance response information. The images and videos corresponding to the maintenance response information are, for example, images and videos showing the location of a filter and how to remove the filter.

[0074] The complaint handling notification device 10C includes an association unit 14 in addition to the components included in the complaint handling notification device 10A. The association unit 14 associates at least one of an image and a video corresponding to the FAQ 16A with the FAQ 16A. The association unit 14 stores a new FAQ (FAQ 16B, described later) associated with at least one of the image and the video in the FAQ DB2.

[0075] Even when appropriate maintenance response information is presented to a user, the user may not understand the presented solution and may be unable to resolve the problem on their own. For example, even when the maintenance response information presented to a user is "Please clean the filter," the user may not know where the filter is located or how to remove it. In this case, the user may be unable to resolve the problem on their own. For this reason, in the third embodiment, the association unit 14 associates at least one of an image and a video with the FAQ 16A to create the FAQ 16B, and presents the maintenance response information to the user, including at least one of the image and the video. This allows the user to understand how to address the maintenance response information and resolve the problem on their own.

[0076] Fig. 8 is a diagram for explaining the configuration of the FAQ stored in the preventive maintenance system according to the third embodiment. Of the components in Fig. 8, those that achieve the same functions as the FAQ 16A according to the first embodiment shown in Fig. 2 are assigned the same reference numerals, and duplicated explanations will be omitted.

[0077] The FAQ 16B of the third embodiment is a graph database configured from categories, symptoms, causes, countermeasures, supplements, etc. In the FAQ 16B, categories are associated with symptoms, symptoms are associated with causes, causes are associated with countermeasures, and countermeasures are associated with supplements.

[0078] The supplementary information is the storage address of the image or video corresponding to the measure. Fig. 8 shows a case where the storage addresses of "filter cleaning image" and "filter cleaning video" are associated with the measure "filter cleaning."

[0079] For example, the associating unit 14 generates an FAQ 16B by associating at least one of an image and a video corresponding to the FAQ 16A shown in Figure 2 with the FAQ 16A. This allows the user to refer to at least one of an image and a video corresponding to the FAQ 16B. The preventive maintenance system 1C may include an FAQ generation device 50, an FAQ generation prompt DB 3, a large-scale language model DB 4, and a maintenance data DB 5.

[0080] As described above, according to the third embodiment, the complaint response notification device 10C provides the user with maintenance response information associated with at least one of an image and a video, allowing the user to easily understand the maintenance response information (maintenance response method). In other words, the complaint response notification device 10C provides the user with easy-to-understand FAQ 16B, allowing the user to easily understand the maintenance response information. As a result, the preventive maintenance system 1C can improve user satisfaction and increase the number of self-solving by users, thereby reducing unnecessary service visits.

[0081] Fourth Embodiment Next, a fourth embodiment will be described with reference to Fig. 9. In the fourth embodiment, at least one of an image and a video corresponding to the FAQ 16B is sent to a terminal of a maintenance service company (a maintenance company terminal 80 described later).

[0082] Fig. 9 is a diagram showing the configuration of a preventive maintenance system according to the fourth embodiment. Of the components in Fig. 9, those that achieve the same functions as those in the preventive maintenance system 1B according to the second embodiment shown in Fig. 4 or the preventive maintenance system 1C according to the third embodiment shown in Fig. 7 are given the same reference numerals, and redundant explanations will be omitted.

[0083] Similar to the preventive maintenance system 1B of the second embodiment, the preventive maintenance system 1D of the fourth embodiment is a system that reduces maintenance services for equipment used by a user. The preventive maintenance system 1D includes a complaint response notification device 10D, air conditioning equipment 20, an information collection device 30, a user terminal 40, an FAQ generation device 50, an FAQ DB2, an FAQ generation prompt DB3, a large-scale language model DB4, a maintenance data DB5, and a maintenance company terminal 80. In this way, the preventive maintenance system 1D includes an image / video DB6 and a maintenance company terminal 80 in addition to the components included in the preventive maintenance system 1B, and includes the complaint response notification device 10D instead of the complaint response notification device 10B.

[0084] The image and video DB 6 may be located in the complaint response notification device 10D. The preventive maintenance system 1D may not include the FAQ generation device 50, the FAQ generation prompt DB 3, the large-scale language model DB 4, and the maintenance data DB 5.

[0085] The complaint response notification device 10D includes the components of the complaint response notification device 10C, as well as a maintenance worker notification unit 15. The maintenance worker notification unit 15 transmits at least one of an image and a video to the maintenance company terminal 80 of the maintenance service company. Specifically, the maintenance worker notification unit 15 acquires the image and video included in the FAQ 16B and sends them to the communication unit 13.

[0086] The maintenance worker notifying unit 15 acquires the maintenance response information associated with at least one of the image and the video from the associating unit 14. The communication unit 13 transmits the image and the video associated with the maintenance response information to the maintenance company terminal 80.

[0087] The maintenance company terminal 80 is connected to the network 9 and transmits and receives information via the network 9. The maintenance company terminal 80 is a terminal used by a maintenance company for the air conditioning equipment 20. Like the user terminal 40, the maintenance company terminal 80 is, for example, a smartphone, a tablet terminal, or a mobile phone. The maintenance company terminal 80 receives and displays maintenance response information from the complaint response notification device 10D.

[0088] The maintenance service provider acquires various knowledge and information, but the acquired knowledge and information varies, and even the maintenance service provider may be unable to deal with an issue that can be addressed, or may take the wrong action. For this reason, the complaint response notification device 10D transmits maintenance response information including at least one of an image and a video to the maintenance service provider's maintenance agent terminal 80 in addition to the processing executed by the complaint response notification device 10C.

[0089] Thus, according to embodiment 4, the preventive maintenance system 1D can present images or videos of maintenance work that are easy to understand even for call center receptionists who have no domain knowledge, new on-site service personnel, and the like, thereby making maintenance work more efficient.

[0090] Fifth Embodiment Next, a fifth embodiment will be described with reference to Fig. 10. In the fifth embodiment, at least one of an image and a video corresponding to the FAQ 16A is automatically generated.

[0091] Fig. 10 is a diagram showing the configuration of a preventive maintenance system according to the fifth embodiment. Of the components in Fig. 10, those that achieve the same functions as those in the preventive maintenance system 1C according to the third embodiment shown in Fig. 7 are given the same reference numerals, and redundant explanations will be omitted.

[0092] Similar to the preventive maintenance system 1C of the third embodiment, the preventive maintenance system 1E of the fifth embodiment is a system that reduces maintenance services for equipment used by users. In addition to the components of the preventive maintenance system 1C, the preventive maintenance system 1E includes an image and video generation device 70, a large-scale language model DB 4, and an image and video generation prompt DB 7. Note that the preventive maintenance system 1E may include a complaint response notification device 10D instead of the complaint response notification device 10C.

[0093] The large-scale language model DB4 of the fifth embodiment stores a large-scale language model (second large-scale language model) that generates images or videos from text. Like the large-scale language model of the second embodiment, the large-scale language model of the fifth embodiment is also a natural language processing model trained using a large amount of text data.

[0094] In the preventive maintenance system 1E, the FAQ DB 2, the image / video DB 6, the large-scale language model DB 4, and the image / video generation prompt DB 7 are connected to the image / video generation device 70. At least one of the large-scale language model DB 4 and the image / video generation prompt DB 7 may be located within the image / video generation device 70. The preventive maintenance system 1E may also include an FAQ generation device 50, an FAQ generation prompt DB 3, and a maintenance data DB 5.

[0095] The image / video generating prompt DB 7 stores a prompt (image / video generating prompt) for generating at least one of an image and a video corresponding to the FAQ 16A to become the FAQ 16B.

[0096] The image and video generating device 70 automatically generates at least one of an image and a video corresponding to the FAQ 16 A using the image and video generating prompt. The image and video generating device 70 has a video generating unit 71, an FAQ data acquiring unit 72, and a communication unit 73.

[0097] The FAQ data acquisition unit 72 acquires FAQ 16A from FAQ DB 2. The video generation unit 71 automatically generates at least one of an image and a video from FAQ 16A by causing the large-scale language model to execute the image / video generation prompt in image / video generation prompt DB 7. That is, the video generation unit 71 uses the large-scale language model to generate at least one of an image and a video to be set from FAQ 16A to FAQ 16B. The video generation unit 71 associates the generated at least one of the image and the video with FAQ 16A to generate FAQ 16B.

[0098] The communication unit 73 receives text data 151A (described later) from an external device via the network 9. The communication unit 73 also receives text data 151B (described later) from the FAQ DB 2, and receives text data 151B from the device information DB 8 via the network 9. The communication unit 73 may also receive text data 151B from the FAQ DB 2 via the complaint response notification device 10C.

[0099] Furthermore, the communication unit 73 transmits the FAQ 16B to the complaint response notification device 10C via the network 9. The communication unit 73 may transmit the FAQ 16B to the FAQ DB 2 via the complaint response notification device 10C, or may transmit the FAQ 16B directly to the FAQ DB 2.

[0100] In the third and fourth embodiments, for example, an image described in an existing manual or the like is associated with FAQ 16A and presented to the user as FAQ 16B, or a video is manually generated and presented to the user as FAQ 16B. The image and video generation device 70 of the fifth embodiment uses a large-scale language model to generate at least one of an image and a video corresponding to FAQ 16A and presents the generated image and video to the user as FAQ 16B. This allows the image and video generation device 70 to reduce the cost and effort required to generate an image or video.

[0101] The large-scale language model used by the image and video generation device 70 of the fifth embodiment is a trained model such as Emu Video (Meta) that generates images or videos from text, and generates images or videos by inputting device information and a countermeasure for FAQ16A. For example, in the case of a device with model name XXXX that requires filter cleaning, the large-scale language model generates a video on how to clean the filter for model name XXXX.

[0102] Here, we will explain a learning device that generates a trained model that generates images or videos from text, and an inference device that generates images or videos from text using the trained model.

[0103] 11 is a diagram illustrating a configuration of a learning device according to the fifth embodiment. The learning device 110 is connected to a trained model storage unit 120. The learning device 110 and the trained model storage unit 120 may be connected via a network 9. The trained model storage unit 120 may be disposed within the learning device 110.

[0104] The learning device 110 includes a data acquisition unit 111 and a model generation unit 112. The data acquisition unit 111 acquires text data 151A from an external device of the learning device 110. The data acquisition unit 111 also acquires image and video data 152A corresponding to the text data 151A from the external device of the learning device 110. The image and video data 152A includes at least one of image and video data. The text data 151A acquired by the data acquisition unit 111 includes expressions included in the device information, expressions included in the countermeasures (maintenance response information) of FAQ 16B, and the like. The data acquisition unit 111 sends the acquired text data 151A and image and video data 152A to the model generation unit 112.

[0105] The model generation unit 112 learns an appropriate image video 152A corresponding to the text data 151A based on learning data generated based on a combination of the text data 151A and the image video 152A sent from the data acquisition unit 111. In other words, the model generation unit 112 learns an appropriate image video 152A when the text data 151A is satisfied based on learning data generated based on a combination of the text data 151A and the image video 152A. That is, the model generation unit 112 generates a trained model 160 that infers an appropriate image video 152A from the text data 151A. Here, the learning data is data in which the text data 151A and the image video 152A are associated with each other.

[0106] The model generation unit 112 can use, as a learning algorithm, known algorithms such as supervised learning, unsupervised learning, reinforcement learning, etc. As an example, a case where a neural network is applied to the learning algorithm used by the model generation unit 112 will be described.

[0107] The model generation unit 112 learns appropriate image video 152A corresponding to text data 151A by so-called supervised learning, for example, according to a neural network model. Here, supervised learning refers to a technique in which data sets (learning data) of inputs and results (labels) are provided to the learning device 110, and the learning device 110 learns features contained in the learning data and infers results from the inputs.

[0108] A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer or two or more layers.

[0109] Fig. 12 is a diagram illustrating a neural network used by the learning device according to the fifth embodiment. For example, in a three-layer neural network as shown in Fig. 12, when multiple inputs are input to the input layer (X1 to X3), the values ​​are multiplied by a weight W1 (shown as w11 to w16 in Fig. 12) and input to the intermediate layer (Y1 to Y2). The result is then further multiplied by a weight W2 (shown as w21 to w26 in Fig. 12) and output from the output layer (Z1 to Z3). This output result varies depending on the values ​​of the weights W1 and W2.

[0110] 12 learns image video 152A corresponding to text data 151A by so-called supervised learning in accordance with learning data generated based on a combination of text data 151A and image video 152A acquired by data acquisition unit 111. In other words, the neural network used by learning device 110 in Fig. 12 learns image video 152A corresponding to text data 151A by so-called supervised learning in accordance with text data 151A and image video 152A generated based on a combination of a first input and a second input (correct answer) acquired by data acquisition unit 111.

[0111] That is, the neural network learns by inputting the first input, text data 151A, and adjusting the weights W1 and W2 so that the result output from the output layer approaches the second input (correct answer).

[0112] In this way, the neural network learns by inputting text data 151A into the input layer and adjusting weights W1 and W2 so that the result output from the output layer approaches image video 152A. By learning the correspondence between text data 151A and image video 152A, the neural network generates trained model 160 that can output appropriate image video 152A when text data 151A is input. In this way, learning device 110 learns trained model 160 that can output correct image video 152A when text data 151A is input.

[0113] By performing the above-described learning, the model generation unit 112 generates and outputs the trained model 160. The trained model storage unit 120 stores the trained model 160 output from the model generation unit 112.

[0114] Next, a processing procedure of the process in which the learning device 110 learns the trained model 160 will be described with reference to Fig. 13. Fig. 13 is a flowchart showing the processing procedure of the learning process executed by the learning device according to the fifth embodiment.

[0115] The data acquisition unit 111 acquires learning data to be used for learning (step S110). Specifically, the data acquisition unit 111 acquires text data 151A and image video 152A.

[0116] Note that data acquisition unit 111 acquires text data 151A and image video 152A simultaneously, but text data 151A and image video 152A may be input in an associated manner. Therefore, data acquisition unit 111 may acquire text data 151A and image video 152A at different times. Data acquisition unit 111 sends text data 151A and image video 152A to model generation unit 112.

[0117] The model generation unit 112 executes a learning process using the text data 151A and the image video 152A (step S120). Specifically, the model generation unit 112 learns the image video 152A corresponding to the text data 151A by so-called supervised learning in accordance with learning data generated based on a combination of the text data 151A and the image video 152A acquired by the data acquisition unit 111, and generates a trained model 160.

[0118] After generating the trained model 160, the model generation unit 112 outputs the trained model 160 to the trained model storage unit 120 (step S130). The trained model storage unit 120 stores the trained model 160 generated by the model generation unit 112.

[0119] 14 is a diagram showing the configuration of an inference device according to the fifth embodiment. The inference device 130 is connected to the trained model storage unit 120. The inference device 130 and the trained model storage unit 120 may be connected via a network 9. The trained model storage unit 120 may be disposed within the inference device 130.

[0120] The inference device 130 has a data acquisition unit 131 and an inference unit 132. The data acquisition unit 131 acquires text data 151B from the FAQ DB2 and the device information DB8. The data acquisition unit 131 acquires text data of countermeasures (maintenance response information) included in the FAQ 16B from the FAQ DB2, and acquires text data of device information corresponding to the countermeasures from the device information DB8. The data acquisition unit 131 acquires a combination of the text data of the countermeasures and the text data of the device information as text data 151B. The text data 151B includes expressions included in the device information, expressions included in the countermeasures (maintenance response information) in the FAQ 16B, and the like. The data acquisition unit 131 sends the acquired text data 151B to the inference unit 132.

[0121] The inference unit 132 receives text data 151B sent from the data acquisition unit 131. The inference unit 132 also reads out the trained model 160 from the trained model storage unit 120. The inference unit 132 uses the trained model 160 to infer an image video 152B corresponding to the text data 151B. That is, the inference unit 132 inputs the text data 151B acquired by the data acquisition unit 131 into the trained model 160, and outputs the appropriate image video 152B inferred from the text data 151B to the image video DB 6.

[0122] Next, a description will be given of the processing procedure of the inference device 130 inferring the image video 152B using the trained model 160. Fig. 15 is a flowchart showing the processing procedure of the inference processing executed by the inference device according to the fifth embodiment.

[0123] The data acquisition unit 131 acquires inference data to be used for inference of the image video 152B (step S210). Specifically, the data acquisition unit 131 acquires text data 151B. The data acquisition unit 131 sends the text data 151B to the inference unit 132. The inference unit 132 acquires the text data 151B from the data acquisition unit 131 and acquires the trained model 160 from the trained model storage unit 120.

[0124] The inference unit 132 inputs the text data 151B into the trained model 160 (step S220) and obtains an appropriate image video 152B corresponding to the text data 151B.

[0125] The inference unit 132 outputs data inferred using the trained model 160 and the text data 151B (step S230). Specifically, the inference unit 132 transmits the appropriate image video 152B obtained by the trained model 160 to the image video DB 6.

[0126] The image / video DB 6 stores the image / video 152B corresponding to the text data 151B (step S240). This allows the complaint handling notification device 10C to acquire the image / video 152B corresponding to the text data 151B from the image / video DB 6.

[0127] At least one of the learning device 110 and the inference device 130 may be located within the complaint response notification device 10C. The trained model storage unit 120 may be located within the complaint response notification device 10C. The inference device 130 may be connected to the image and video DB 6 via the network 9 or a network other than the network 9, or may be directly connected to the image and video DB 6. At least one of the learning device 110, the inference device 130, and the trained model storage unit 120 may reside on a cloud server.

[0128] In the fifth embodiment, an example has been described in which the model generation unit 112 uses a supervised learning algorithm as a learning algorithm, but the learning algorithm used by the model generation unit 112 is not limited to a supervised learning algorithm. The model generation unit 112 can also apply a reinforcement learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm, or the like, in addition to a supervised learning algorithm.

[0129] Deep learning, which learns to extract features themselves, can also be used as the learning algorithm of the model generation unit 112. The model generation unit 112 may also perform machine learning according to other known methods, such as genetic programming, functional logic programming, or support vector machines.

[0130] The inference device 130 may be applied to the estimation unit 12 described in the first embodiment and the like. In this case, the inference device 130 infers a complaint corresponding to the equipment information using a trained model for inferring a complaint from the equipment information. In this case, the trained model used by the inference device 130 is generated by a learning device. The learning device generates a trained model for inferring complaints that may be entered by users of the air conditioning equipment 20 from the equipment information of the air conditioning equipment 20 using training data including the equipment information and the complaints corresponding to the equipment information. The complaints corresponding to abnormal states in the training data are complaints in FAQ 16A. The learning device may generate a trained model for inferring complaints that may be entered by users of the air conditioning equipment 20 from the abnormal state of the air conditioning equipment 20 using training data including the abnormal state and the complaint corresponding to the abnormal state.

[0131] Thus, according to the fifth embodiment, the image / video generating device 70 automatically generates at least one of an image and a video corresponding to the FAQ 16A, thereby reducing the cost and effort required to generate the image or video.

[0132] Next, we will explain the hardware configurations of the complaint response notification devices 10A, 10C, and 10D, the information collection device 30, the image and video generation device 70, the learning device 110, and the inference device 130. Note that the complaint response notification devices 10A, 10C, and 10D, the information collection device 30, the image and video generation device 70, the learning device 110, and the inference device 130 have similar hardware configurations, so here we will explain the hardware configuration of the complaint response notification device 10C according to the fifth embodiment.

[0133] The complaint handling notification device 10C is realized by a processing circuit. This processing circuit may be a processor and memory that executes a program stored in memory, or may be dedicated hardware. The processing circuit is also called a control circuit.

[0134] 16 is a diagram showing an example of the configuration of a processing circuit provided in a complaint handling notification device according to the fifth embodiment, when the processing circuit is realized by a processor and a memory. The processing circuit 90 shown in FIG. 16 is a control circuit, and includes a processor 91 and a memory 92. When the processing circuit 90 is configured with the processor 91 and the memory 92, each function of the processing circuit 90 is realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 92.

[0135] In the processing circuit 90, each function is realized by the processor 91 reading and executing a program stored in the memory 92. That is, the processing circuit 90 includes a memory 92 for storing a complaint response notification program that results in the processing of the complaint response notification device 10C. This complaint response notification program can also be said to be a program that causes the complaint response notification device 10C to execute each function realized by the processing circuit 90. This complaint response notification program may be provided by a storage medium on which the complaint response notification program is stored, or by other means such as a communication medium.

[0136] The complaint handling notification program executed by the complaint handling notification device 10C has a modular configuration including an estimation unit 12, a communication unit 13, and an association unit 14, which are loaded onto the main memory device and generated on the main memory device.

[0137] Here, the processor 91 is, for example, a CPU (Central Processing Unit), a processing device, an arithmetic unit, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor), etc. The memory 92 is, for example, a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM), or an EEPROM (Electrically EPROM), a magnetic disk, a flexible disk, an optical disk, a compact disk, a minidisk, or a DVD (Digital Versatile Disc).

[0138] Fig. 17 is a diagram showing an example of a processing circuit when the processing circuit provided in the complaint handling notification device according to the fifth embodiment is configured with dedicated hardware. The processing circuit 93 shown in Fig. 17 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof.

[0139] The processing circuits 90 and 93 may be partially implemented by dedicated hardware and partially implemented by software or firmware. In this way, the processing circuits 90 and 93 can realize the above-described functions by dedicated hardware, software, firmware, or a combination of these.

[0140] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or different embodiments may be combined with each other. It is also possible to omit or modify parts of the configurations as long as they do not deviate from the gist of the invention.

[0141] 1A to 1E Preventive maintenance system, 2 FAQ DB, 3 FAQ generation prompt DB, 4 Large-scale language model DB, 5 Maintenance data DB, 6 Image video DB, 7 Image video generation prompt DB, 8 Equipment information DB, 9 Network, 10A, 10C, 10D Complaint response notification device, 11 FAQ storage unit, 12 Estimation unit, 13, 33, 53, 73 Communication unit, 14 Association unit, 15 Maintenance worker notification unit, 16A, 16B FAQ, 20 Air conditioning equipment, 30 Information collection device, 31 Information acquisition unit, 32 Equipment information storage unit, 40 User terminal, 50 FAQ generation device, 51 Maintenance response estimation unit, 52 Maintenance data acquisition unit, 54 FAQ generation prompt, 55 Maintenance data, 70 Image video generation device, 71 Video generation unit, 72 FAQ data acquisition unit, 80 Maintenance company terminal, 90, 93 processing circuit, 91 processor, 92 memory, 110 learning device, 111, 131 data acquisition unit, 112 model generation unit, 120 trained model storage unit, 130 inference device, 132 inference unit, 151A, 151B text data, 152A, 152B image video, 160 trained model.

Claims

1. An information collection device that collects equipment information, which is information about the status of equipment, A complaint response notification device that transmits maintenance response information to a user terminal, based on the aforementioned equipment information, which associates complaints unrelated to equipment failure with responses to those complaints. Equipped with, The aforementioned complaint handling notification device is An estimation unit estimates the complaint based on the aforementioned equipment information and reads maintenance response information corresponding to the estimated complaint from among multiple maintenance response information items. A communication unit that transmits the retrieved maintenance information to the user terminal, A preventive maintenance system having the following features.

2. The device further comprises a maintenance information generation device that generates maintenance response information corresponding to maintenance data from maintenance data indicating the work performed by maintenance workers in response to abnormal conditions of the equipment, using a first large-scale language model trained on text data. The preventive maintenance system according to claim 1.

3. The complaint response notification device further includes an association unit that associates the maintenance response information with at least one of an image and a video corresponding to the maintenance response information. The communication unit transmits at least one of the image and video corresponding to the maintenance information to the user terminal. The preventive maintenance system according to claim 1 or 2.

4. The communication unit transmits at least one of the images and videos corresponding to the maintenance response information to a maintenance service provider's terminal, which is the terminal of the maintenance service provider. The preventive maintenance system according to claim 3.

5. The device further comprises an image and video generation device that generates at least one of an image and a video corresponding to the maintenance information from the maintenance information using a second large-scale language model trained on text data. The preventive maintenance system according to claim 1.

6. The system further comprises an inference device that infers the complaint from the equipment information using a trained model for inferring the complaint from the equipment information. The preventive maintenance system according to claim 1.

7. The learning device further includes a device that acquires the aforementioned device information and the aforementioned complaints as training data, and generates the trained model based on the training data. The preventive maintenance system according to claim 6.