Proposal assistance device, proposal assistance method, and program

Abstract

A proposal assistance device (100) comprises: an IoT information acquisition unit (101) that acquires, from an apparatus to be monitored, IoT information including an operation state of the apparatus and environment information indicating the environment of an installation location of the apparatus; a learning unit that learns a relationship between the time series of the IoT information, a failure state including the presence / absence of a failure and a failure content of the apparatus, and a correspondence history indicating that either a repair or an update has been performed to the apparatus; and an inference unit that, on the basis of the time series including the IoT information newly acquired by the IoT information acquisition unit 101, infers a failure that may occur within a set period to the apparatus that has outputted the IoT information, and outputs a proposal including at least one of a repair service for the apparatus and a renewal sale for updating the apparatus.

Description

Proposal support device, proposal support method, and program

[0001] The present disclosure relates to a proposal support device, a proposal support method, and a program.

[0002] There is known a technology for assisting repair work by estimating the timing of occurrence of an abnormality or failure of a device. For example, Patent Literature 1 discloses a repair support server that estimates the timing of occurrence of an abnormality of a specific function based on inspection result information output from a multifunction peripheral device, and if a predetermined period of time has passed before the estimated occurrence of the abnormality, outputs arrangement information including parts to be replaced in repair work corresponding to the abnormality that is predicted to occur.

[0003] JP 2012-181594 A

[0004] This repair support server estimates the time of occurrence of an abnormality based on inspection result information from a multifunction peripheral device that has a periodic inspection function that periodically inspects for deterioration of specific components and degradation of specific functions. Therefore, there is a problem that it cannot be applied to devices that do not have such a periodic inspection function. Furthermore, the repair support server disclosed in Patent Document 1 does not take into account so-called renewal sales, which are the replacement of existing devices with new devices, so there is room for improvement in terms of proposing appropriate measures for potential malfunctions.

[0005] The present disclosure has been made in consideration of the above-described circumstances, and aims to provide a proposal support device, a proposal support method, and a program that are capable of proposing more appropriate responses to possible failures.

[0006] In order to achieve the above object, the proposal support device according to the present disclosure includes an IoT information acquisition unit that acquires IoT information from a monitored device, including the operating status of the device and environmental information indicating the environment where the device is installed; a learning unit that learns the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history indicating whether or not the device has been repaired or updated; and an inference unit that infers failures that may occur within a set period of time for the device that output the IoT information, based on the time series including the IoT information newly acquired by the IoT information acquisition unit, and outputs a proposal including at least one of a repair service for the device and a renewal sale to update the device.

[0007] According to the present disclosure, the relationship between the time series of IoT information output from a device, the device's failure status, and the history of responses taken on the device is learned, and possible failures that may occur within a set period are inferred, and a proposal including at least one of repair service and renewal sales is output. As a result, it is possible to propose more appropriate responses to possible failures.

[0008] 1. Block diagram showing the functional configuration of a proposal support device according to embodiment 1. Flowchart showing information acquisition processing by the IoT information acquisition unit shown in FIG. 1. Diagram showing an example of IoT information acquired by the IoT information acquisition unit shown in FIG. 1. Diagram showing an example of a customer master. Diagram showing an example of a model name master. Flowchart showing IoT information storage processing by the IoT information acquisition unit shown in FIG. 1. Flowchart showing information transmission processing by the IoT information output unit and the response history output unit shown in FIG. 1. Block diagram showing the functional configuration of a learning device shown in FIG. 1. Flowchart showing learning processing by the learning device shown in FIG. 1. Block diagram showing the functional configuration of an inference device shown in FIG. 1. Flowchart showing inference information acquisition processing by the data acquisition unit shown in FIG. 10. Flowchart showing input data generation processing by the data preprocessing unit shown in FIG. 10. 1. Flowchart showing inference processing by the unit. FIG. 1 shows an example of production management information stored in the production management database shown in FIG. 1. Flowchart showing renewal sales proposal generation processing by the renewal sales proposal generation unit shown in FIG. 1. FIG. 1 shows an example of parts management information stored in the parts management database shown in FIG. 1. Flowchart showing repair service proposal generation processing by the repair service proposal generation unit shown in FIG. 1. Flowchart showing display processing by the display unit shown in FIG. 1. Flowchart showing response history storage processing by the response history input unit shown in FIG. 1. Block diagram showing an example of the physical configuration of the proposal support device. Flowchart showing response proposal processing by the proposal support device. FIG. 1 shows an example of renewal sales proposal information generated by the proposal support device. FIG.

[0009] Hereinafter, a proposal support device, a proposal support method, and a program according to embodiments of the present disclosure will be described with reference to the drawings. Note that the same reference numerals are used to denote the same or corresponding parts in the drawings.

[0010] As illustrated in FIG. 1 , a proposal support device 100 according to this embodiment is communicatively connected to a plurality of air conditioning equipment 200, which are equipment to be monitored. The air conditioning equipment 200 has various sensors that detect the operating status of the air conditioning equipment 200, environmental information related to the environment of the installation location of the air conditioning equipment 200, and the like. The air conditioning equipment 200 transmits IoT information, including information stored in the air conditioning equipment 200 and information acquired by the air conditioning equipment 200 using sensors, to the proposal support device 100 at preset times, such as daily or weekly. The air conditioning equipment 200 may have a function to notify the outside of the occurrence of an abnormality, in which case it may notify the proposal support device 100 of an error code indicating the abnormal state.

[0011] The proposal support device 100 applies learning information based on IoT information acquired from the air conditioning equipment 200 to the trained learning model to infer failures within a preset period such as one month, three months, six months, or one year, and outputs response proposals including repairs to the air conditioning equipment 200 and renewal sales indicating the replacement of the currently used air conditioning equipment 200 with a new air conditioning equipment 200.

[0012] The proposal support device 100 according to this embodiment includes an IoT information acquisition unit 101 that acquires various information including IoT information of the air conditioning equipment 200, an IoT information database 102 that stores the IoT information acquired by the IoT information acquisition unit 101, a response history database 103 that stores response history information indicating a history of responses that have been implemented in the past for the air conditioning equipment 200, an IoT information output unit 104 that outputs the IoT information stored in the IoT information database 102 to a learning device 106 or an inference device 107, a response history output unit 105 that outputs the response history information to the learning device 106, the learning device 106 that generates a trained model, and an inference device 107 that infers a failure of the air conditioning equipment 200 within a set period and outputs a response proposal to be implemented for the air conditioning equipment 200. The system comprises a device 107, a production management database 108 that stores production management information for each model name of the air conditioning equipment 200, a renewal sales proposal generation unit 109 that generates renewal sales proposal content by utilizing the production management database 108, a parts management database 110 that stores production and procurement information for parts that the air conditioning equipment 200 has, a repair service proposal generation unit 111 that generates proposal content for repair services by utilizing the parts management database 110, a display unit 112 that displays the proposal content generated by the renewal sales proposal generation unit 109 and the repair service proposal generation unit 111, and a response information input unit 113 that accepts input of the response history performed on the air conditioning equipment 200 and stores it in the response history database 103.

[0013] The IoT information acquisition unit 101 executes an information acquisition process to acquire IoT information of the air conditioning equipment 200, customer information indicating information about the customer of the air conditioning equipment 200, and model name information indicating information about the model name of the air conditioning equipment 200.

[0014] 2 shows a flowchart of information acquisition processing by the IoT information acquisition unit 101. As shown in the figure, the IoT information acquisition unit 101 acquires IoT information, such as that shown in FIG. 3, from each air conditioning device 200 that has been delivered to a customer at regular intervals, such as daily or weekly intervals (step S11). Specifically, as shown in the figure, the IoT information includes model information, such as the model name, software version, and serial number of the air conditioning device 200; date information, such as the manufacturing date, installation date, durability years, and most recent maintenance date of the air conditioning device 200; environmental information, such as maximum, minimum, and average values ​​of temperature and humidity, the installation location of the outdoor unit, and the number of connected indoor units; and operation information, such as the average usage time of the air conditioning device 200, total usage time, and the number of times each function, including heating, cooling, and dehumidification, has been used.

[0015] Returning to Figure 2, next, the IoT information acquisition unit 101 reads the customer master 20 illustrated in Figure 4, which is stored in the memory unit of the proposal support device 100, and acquires customer information including information such as the customer name, target country, target region, and address of the air conditioning equipment 200 that output the IoT information acquired in step S11 (step S12).

[0016] Returning to Figure 2, next, the IoT information acquisition unit 101 reads the model name master 21 illustrated in Figure 5, which is stored in the memory unit of the proposal support device 100, and acquires model name information including information such as the model series, category, and type of refrigerant of the air conditioning equipment 200 that output the IoT information acquired in step S11 (step S13).

[0017] Returning to FIG. 2, next, the IoT information acquisition unit 101 stores the various pieces of information acquired in steps S11 to S13 in a temporary storage area of ​​the memory provided in the proposal support device 100 (step S14).

[0018] Returning to FIG. 1 , the IoT information database 102 stores the IoT information acquired by the IoT information acquisition unit 101 .

[0019] 6 shows a flowchart of an IoT information storage process in which the IoT information acquisition unit 101 stores IoT information in the IoT information database 102. First, the IoT information acquisition unit 101 reads out the IoT information stored in the temporary storage area of ​​the memory by the information acquisition process illustrated in FIG. 2 (step S21), and acquires the current date by referring to the calendar master 22 that manages dates (step S22).

[0020] Next, the IoT information acquisition unit 101 acquires the latest IoT information of the target air conditioning equipment 200 stored in the IoT information database 102, and determines whether there is a difference between the IoT information read in step S21 and the latest IoT information (step S23). If the IoT information acquisition unit 101 determines that there is a difference (step S23; Yes), it associates the IoT information read in step S21 with the date acquired in step S22 and stores it in the IoT information database 102 (step S24).

[0021] On the other hand, if the IoT information acquisition unit 101 determines that there is no difference (step S23; No), it ends the process without storing the IoT information read in step S21 in the database 102. In this way, the IoT information database 102 stores the IoT information in association with the acquisition date, so that it is possible to check the content of the IoT information by date and to check changes in the IoT information over time.

[0022] 1 , the response history database 103 stores response history information indicating the history of responses previously performed on each air conditioning equipment 200. Specifically, the response history information includes information such as a serial number, which is information that uniquely identifies the air conditioning equipment 200, the date on which the response work was performed, the response content indicating the details of the repair or renewal sale performed on each air conditioning equipment 200, the failure occurrence date and time indicating the date and time when the failure occurred, and the failure content indicating the content of the failure, including the location of the failure.

[0023] The response details of the response history information may include information identifying parts replaced in the repair work, or may include the model name of the air conditioning equipment 200 after renewal sales, i.e., after updating. The date and time of the failure may be input based on the date and time an error code was output from the air conditioning equipment 200, or the date and time the failure was discovered by the worker who repaired the air conditioning equipment 200 may be input. The details of the failure may be input based on error information output from the air conditioning equipment 200, or may be input by the worker who repaired the air conditioning equipment 200. Note that if repairs or renewal sales were performed before the air conditioning equipment 200 failed, information does not need to be input for the date and time of the failure and the details of the failure in the response history information.

[0024] When the learning process is executed by the learning device 106, the IoT information output unit 104 transmits the IoT information stored in the IoT information database 102 to the learning device 106. In addition, when the inference process is executed by the inference device 107, the IoT information output unit 104 transmits the IoT information stored in the IoT information database 102, device information including customer information and model name information to the inference device 107.

[0025] The response history output unit 105 outputs the response history information stored in the response history database 103 to the learning device 106 when the learning process is executed by the learning device 106 .

[0026] 7 is a flowchart of an information transmission process in which the IoT information output unit 104 and the response history output unit 105 transmit IoT information and response history information for generating learning data to be used for learning to the learning device 106. This information transmission process is performed, for example, when new response history information is stored in the response history database 103, in order to transmit this response history information and the IoT information of the corresponding air conditioning equipment 200 to the learning device 106.

[0027] As shown in the figure, first, the response history output unit 105 reads newly stored response history information from the response history database 103 (step S31), and transmits it to the learning device 106 (step S32).

[0028] Next, the IoT information output unit 104 acquires the IoT information of the air conditioning equipment 200 from the IoT information database 102 using the serial number of the air conditioning equipment 200 included in the response history information as a key (step S33). Specifically, the IoT information database 102 stores IoT information for each air conditioning equipment 200 for each date on which the IoT information was stored. Therefore, the IoT information output unit 104 acquires IoT information for all dates including the serial number of the air conditioning equipment 200 to be processed and transmits the acquired IoT information to the learning device 106 (step S34). Note that if the period of the IoT information to be acquired is set in advance, the IoT information output unit 104 may extract IoT information for a period, for example, within six months or one year from the date of response implementation, and transmit the IoT information to the learning device 106.

[0029] 1 , the learning device 106 learns the relationship between the IoT information, the failure status of the air conditioning equipment 200, and the response to the air conditioning equipment. Specifically, as illustrated in FIG. 8 , the learning device 106 includes a data acquisition unit 1061, a data preprocessing unit 1062, a learning model generation unit 1063, and a learned model accumulation unit 1064. Note that the learning device 106 is an example of a learning unit.

[0030] The data acquisition unit 1061 acquires IoT information and response history information used to generate learning data. The data acquisition unit 1061 is an example of a learning information acquisition unit.

[0031] The data pre-processing unit 1062 generates learning data based on the IoT information and response history information acquired by the data acquisition unit 1061. Specifically, the data pre-processing unit 1062 generates time-series data for each air conditioning equipment 200 by arranging the IoT information in chronological order of acquisition date and compiling it into a single piece of data, and generates learning data by associating the time-series data with the response history information of each air conditioning equipment 200.

[0032] The learning model generation unit 1063 generates a trained model based on the training data generated by the data preprocessing unit 1062. Specifically, the learning model generation unit 1063 provides the training data to the training model to generate a trained training model. The learning algorithm used by the learning model generation unit 1063 can be a known supervised learning algorithm. Supervised learning refers to a method of providing a combination of input and result data, learning features in the training data, and inferring a result from the input. The training data includes time-series data of IoT information, failure conditions including whether or not a failure has occurred and the location of the failure, and measures taken for each air conditioning equipment 200. Therefore, a trained model is generated that infers the probability of occurrence of a failure that may occur within a set period and the location of the failure based on changes in the time series of the IoT information, and outputs a proposed measure to be taken for the air conditioning equipment 200.

[0033] The trained model accumulation unit 1064 stores the trained model generated by the trained model generation unit 1063. Note that, for example, a different trained model may be generated for each model name of the air conditioning equipment 200, and in this case, the trained model accumulation unit 1064 may store the multiple trained models generated.

[0034] 9 shows the flow of the learning process by the learning device 106. First, the data acquisition unit 1061 acquires the IoT information and the response history information transmitted by the IoT information output unit 104 and the response history output unit 105 through the information transmission process illustrated in FIG.

[0035] Next, the data preprocessing unit 1062 generates time-series data by arranging the IoT information in chronological order of acquisition date and consolidating it into a single data set, and generates training data by associating the time-series data and response history information with the serial number of the air conditioning equipment 200 as a key. Next, the training model generation unit 1063 executes a training process based on the training data, which is a combination of the time-series data of the IoT information and the response history information generated by the data preprocessing unit 1062 (step S42), to generate a trained model. The generated trained model is then stored in the trained model accumulation unit 1064 (step S43).

[0036] 1 , the inference device 107 outputs the failure probability and failure location within a set period, and a countermeasure plan, using device information including the IoT information, customer information, and model name information of the air conditioning equipment 200 acquired by the IoT information acquisition unit 101, and the trained model generated by the learning device 106. Specifically, as illustrated in FIG. 10 , the inference device 107 includes a data acquisition unit 1071, a data preprocessing unit 1072, and an inference unit 1073.

[0037] The data acquisition unit 1071 executes an inference information acquisition process to acquire information used to generate input data for inference. The flow of the inference information acquisition process by the data acquisition unit 1071 is shown in Fig. 11. Specifically, when new IoT information is stored in the IoT information database 102 by the IoT information process illustrated in Fig. 6, the data acquisition unit 1071 acquires, from the IoT information database 102, the stored new IoT information and past IoT information of the air conditioning equipment 200 that is the sender of the IoT information (step S51).

[0038] Next, the data acquisition unit 1071 searches the response history database 103 and acquires past response history information for the air conditioning equipment 200 to be processed (step S52). Note that if there is no response history information for this air conditioning equipment 200, the process proceeds to step S53 without performing any processing.

[0039] Next, the data acquisition unit 1071 references the model name master 21 illustrated in FIG. 5 to acquire model name information including the model series, category, type of refrigerant, etc. of the air conditioning equipment 200 to be processed (step S53), and references the customer master 20 illustrated in FIG. 4 to acquire customer information including the customer name, target country, target region, and address of the air conditioning equipment 200 to be processed (step S54).

[0040] Next, the data acquisition unit 1071 connects the pieces of information acquired in steps S51 to S54 (step S55), and stores the connected data in a temporary storage area of ​​the memory (step S56).

[0041] Returning to Fig. 10, the data preprocessing unit 1072 executes input data generation processing to generate input data for inference based on data stored in the temporary storage area by the data acquisition unit 1061. The data preprocessing unit 1072 is an example of an information-for-inference generation unit. Fig. 12 shows the flow of the input data generation processing by the data preprocessing unit 1072. First, the data preprocessing unit 1072 acquires data stored by the information-for-inference acquisition processing illustrated in Fig. 11 from the temporary storage area of ​​the memory (step S61).

[0042] Next, the data preprocessing unit 1072 processes the information acquired in step S61 (step S62) to generate input data for inference. Specifically, the data preprocessing unit 1072 generates input data by sorting the IoT information in order of acquisition date to generate time-series data and by performing processing such as removing outliers. Next, the data preprocessing unit 1072 stores the generated input data in a temporary area of ​​memory (step S63).

[0043] 10 , the inference unit 1073 applies the input data generated by the data preprocessing unit 1072 to the learned model to infer the failure probability, failure location, and response plan for the air conditioning equipment 200 within a set period. That is, the inference unit 1073 receives input data including time-series data of IoT information, response history information, model name information, and customer information for the air conditioning equipment 200 to be processed, and outputs the failure probability, failure location, and response plan for the set period. Note that the inference unit 1073 is an example of an inference result generation unit.

[0044] In addition, the inference unit 1073 is not limited to outputting inference results using a trained model generated by the model generation unit 1062 of the learning device 106, but may also obtain a trained model from outside the proposal support device 100 and perform inference processing based on this trained model.

[0045] 13 shows the flow of the inference process by the inference unit 1073. First, the inference unit 1073 acquires input data for inference stored in a temporary storage area through the input data generation process illustrated in FIG. 12 (step S71).

[0046] Next, the inference unit 1073 inputs the input data into the trained model registered in the trained model accumulation unit 1064 (step S72), and obtains an inference result output from the trained model. Specifically, the inference result includes the failure probability within a set period, the failure location, and at least one of the countermeasures of repair and renewal sales.

[0047] Next, the inference unit 1073 outputs the inference result to the renewal sales proposal generation unit 109 and the repair service proposal generation unit 111 (step S73).

[0048] 1, the production management database 108 stores production management information for each model name of the air conditioning equipment 200. As shown in Fig. 14, the production management information includes, for each model name, information on the inventory quantity indicating inventory information, and the planned production date, planned production quantity, and possible shipping date indicating production plan information, and an alternative model name.

[0049] 1, the renewal sales proposal generation unit 109 executes a renewal sales proposal generation process that generates renewal sales proposal content by utilizing the production management database 108 based on the inference result by the inference device 107. The flow of the renewal sales proposal generation process is shown in FIG.

[0050] First, the renewal sales proposal generation unit 109 obtains an inference result including a renewal sales proposal from the inference device 107 (step S81), and obtains IoT information, model name information, and customer information of the air conditioning equipment 200 to be processed from the IoT information database 102 (step S82).

[0051] Next, the renewal sales proposal generation unit 109 obtains from the production management database 108 the inventory quantity, planned production date, planned production quantity, available shipping date, and alternative model name of the same model name as the air conditioning equipment 200 being processed (step S83). Next, the renewal sales proposal generation unit 109 generates renewal sales proposal information by linking the failure probability and failure location included in the inference result obtained in step S81 with the IoT information, model name information, and customer information obtained in step S82, and the inventory quantity, planned production date, planned production quantity, available shipping date, and alternative model name obtained in step S83 (step S84). Note that if production of the air conditioning equipment 200 with the same model name as the processing target has ended or if the available shipping date is after the period in which a failure is expected to occur, the inventory quantity, planned production date, and planned production quantity of the alternative model name may be included in the renewal sales proposal information.

[0052] Next, the renewal sales proposal generating unit 109 stores the generated renewal sales proposal information in a temporary storage area of ​​the memory (step S85).

[0053] 1, the parts management database 110 stores parts management information indicating production and procurement information for parts included in the air conditioning equipment 200. Specifically, as shown in Fig. 16, the parts management information includes information such as the work content of repair services that can be performed for each model name, the parts required for each repair service and their inventory quantities, the dates when parts that are out of stock can be produced and procured, and the dates when each repair service can be performed.

[0054] The repair service proposal generation unit 111 executes a repair service proposal generation process that derives repair service proposal content by utilizing the parts management database 110 based on the inference result by the inference device 107. The flow of the repair service proposal generation process is shown in FIG.

[0055] First, the repair service proposal generation unit 111 acquires the inference result from the inference device 107 (step S91), and acquires the IoT information, model name information, and customer information of the air conditioning equipment 200 to be processed from the IoT information database 102 (step S92).

[0056] Next, the repair service proposal generation unit 111 obtains from the parts management database 110 the parts and number of parts required for the repair service included in the inference result, the inventory quantity of the parts, the production / procurement date of the parts, and the date on which the repair service can be carried out (step S93).

[0057] Next, the repair service proposal generation unit 111 generates repair service proposal information by linking the failure probability contained in the inference result obtained in step S91 with the IoT information, model name information, and customer information obtained in step S92, and the repair service work content, parts and number of parts required for the repair service, the inventory number of the parts and the date when production / procurement is possible, and the time when the repair service can be performed obtained in step S93 (step S94).

[0058] Next, the repair service proposal generating unit 111 stores the generated repair service proposal information in a temporary storage area of ​​the memory (step S95).

[0059] 1, the display unit 112 executes a display process for displaying the renewal sales proposal information generated by the renewal sales proposal generation unit 109 and the repair service proposal information generated by the repair service proposal generation unit 111. The flow of the display process is shown in FIG.

[0060] The display unit 112 acquires the renewal sales proposal information stored in the temporary storage area of ​​memory by the renewal sales proposal generation unit 109 (step S101), and acquires the repair service proposal information stored in the temporary storage area of ​​memory by the repair service proposal generation unit 111 (step S102). The display unit 112 displays the renewal sales proposal information and repair service proposal information acquired in steps S101 and S102 on the screen (step S103). Note that if the inference result of the inference device 107 includes only a repair or renewal sales solution, the renewal sales proposal generation unit 109 and the repair service proposal generation unit 111 store only either the renewal sales proposal information or the repair service proposal information in the temporary storage area of ​​memory. In this case, the display unit 112 only needs to display one of the proposal information stored in the temporary storage area.

[0061] 1 , the response information input unit 113 receives input of response history information and executes a response history storage process to store the input response history information in the response history database 103. The response history information is input, for example, by the administrator of the proposal support device 100 or the worker who performed the response work. The flow of the response history storage process is shown in FIG. 19 .

[0062] The response information input unit 113 accepts the selection of any one of the proposal information from the renewal sales proposal information and the repair service proposal information displayed on the display unit 112 (step S111). Note that the selection of the proposal information may be performed by the manager of the proposal support device 100 or the worker in charge before the response, or may be performed after the response.

[0063] When any of the proposed information is selected, the corresponding information input unit 113 refers to the IoT information database 102 and acquires the IoT information, model name information, and customer information of the air conditioning equipment 200 for which the proposed information was selected (step S112).

[0064] Next, the response information input unit 113 stores the contents of the proposal information selected in step S111 and the IoT information, model name information, and customer information acquired in step S112 in the response history database 103 (step S113).

[0065] The proposal support device 100 having the functional configuration described above physically comprises, as shown in FIG. 20 , a CPU (Central Processing Unit) 11 that executes processing according to a program, a RAM (Random Access Memory) 12 which is a volatile memory, a ROM (Read Only Memory) 13 which is a non-volatile memory, a storage unit 14 that stores data, an input unit 15 that accepts input of information, a display unit 16 that visualizes and displays information, and a communication unit 17 that transmits and receives information, all of which are connected via an internal bus 99.

[0066] The CPU 11 executes various processes by reading out the programs stored in the storage unit 14 into the RAM 12 and executing them. The CPU 11 executes processes by an IoT information acquisition unit 101, an IoT information output unit 104, a response history output unit 105, a learning device 106, an inference device 107, a renewal sales proposal generation unit 109, a repair service proposal generation unit 111, and a display unit 112 as main functions provided by the programs.

[0067] The RAM 12 is used as a work area for the CPU 11. The ROM 13 stores the control program executed by the CPU 11, a BIOS (Basic Input Output System), and the like.

[0068] The storage unit 14 includes a hard disk drive, stores programs executed by the CPU 11, and stores various data used when the programs are executed. The storage unit 14 functions as an IoT information database 102, a response history database 103, a production management database 108, and a parts management database 110.

[0069] The input unit 15 is a user interface including a keyboard, a mouse, a communication device, etc. The input unit 15 functions as a correspondence information input unit 113 .

[0070] The display unit 16 is a display device that visualizes and displays information, such as a liquid crystal display, an organic electroluminescence (EL) display, etc. The display unit 16 functions as a display unit 112.

[0071] The communication unit 17 is a network termination device or a wireless communication device that connects to a network, and a serial interface or a LAN (Local Area Network) interface that connects to them.

[0072] Next, the operation of the response proposal process performed by the proposal support device 100 will be described with reference to FIG.

[0073] The trained model accumulation unit 1064 of the proposal support device 100 stores in advance trained models generated by the training process illustrated in FIG. 9 .

[0074] When the IoT information illustrated in FIG. 3 is transmitted from the air conditioning equipment 200, the proposal support device 100 starts the response proposal process.

[0075] In step S121, the IoT information acquisition unit 101 acquires the IoT information transmitted from the air conditioning equipment 200 through the information acquisition process illustrated in FIG. 2 and the IoT information storage process illustrated in FIG. 6, and stores the IoT information in the IoT information database 102.

[0076] In step S122, the data acquisition unit 1071 of the inference device 107 acquires, by acquisition date, the IoT information stored in step S121 and the past IoT information of the air conditioning equipment 200 to be processed from the IoT information database 102 through the inference information acquisition process illustrated in Fig. 11. The data acquisition unit 1071 also references the response history database 103, the customer master, and the model name master to acquire the response history information, customer information, and model name information of the air conditioning equipment 200 to be processed.

[0077] Next, the data preprocessing unit 1072 generates time-series data by sorting the IoT information in order of acquisition date using the input data generation process illustrated in Figure 12, and generates input data for inference that associates the time-series data with response history information, customer information, and model name information.

[0078] In step S123, the inference unit 1073 outputs an inference result including the failure probability within a set period of the air conditioning equipment 200 being processed, the failure location, and a countermeasure, using the inference process exemplified in Fig. 13. For example, the inference unit 1073 outputs an inference result that the failure probability within one year of the air conditioning equipment 200 being processed is 85%, the failure location is the unit, and the countermeasure is renewal sales indicating an update of the air conditioning equipment 200 or replacement of the unit.

[0079] In step S124, the inference unit 1073 determines whether or not renewal sales are included in the response plans resulting from the inference. If the inference unit 1073 determines that renewal sales are included in the response plans resulting from the inference (step S124; Yes), it notifies the renewal sales proposal generation unit 109 of this fact and proceeds to step S125. On the other hand, if the inference unit 1073 determines that renewal sales are not included in the response plans resulting from the inference (step S124; No), it proceeds to step S126.

[0080] In step S125, the renewal sales proposal generation unit 109 derives a renewal sales proposal by the renewal sales proposal generation process illustrated in Fig. 15. Specifically, the renewal sales proposal generation unit 109 references the production management database 108 illustrated in Fig. 14, and generates renewal sales proposal information taking into account the status of the available shipping date of the same model name as the air conditioning equipment 200 being processed or an alternative model name. Specifically, the renewal sales proposal generation unit 109 generates renewal sales proposal information including the model name, available shipping date, inventory quantity, planned production date, and planned production quantity for which renewal sales are being proposed, as illustrated in Fig. 22.

[0081] In step S126, the inference unit 1073 determines whether the countermeasures resulting from the inference include repair. If the inference unit 1073 determines that the countermeasures resulting from the inference include repair (step S126; Yes), it notifies the repair service proposal generation unit 111 of this fact and proceeds to step S127. On the other hand, if the inference unit 1073 determines that the countermeasures resulting from the inference do not include repair (step S126; No), it proceeds to step S128.

[0082] In step S127, the repair service proposal generator 111 derives a repair service proposal by the repair service proposal generation process illustrated in Fig. 17. Specifically, the repair service proposal generator 111 references the parts management database 110 illustrated in Fig. 16 and generates repair service proposal information by taking into account the possible dates for the repair work output by the inference unit 1073. Specifically, the repair service proposal generator 111 generates repair service proposal information including the work content of the proposed repair service, the possible dates for the work, the necessary parts, the inventory quantity of each part, and the production and procurement dates, as illustrated in Fig. 23.

[0083] 21 , in step S128, the display unit 112 displays the response proposals derived by the renewal sales proposal generation unit 109 and the repair service proposal generation unit 111 through the display processing illustrated in Fig. 18. Note that the display unit 112 not only displays the renewal sales proposal information illustrated in Fig. 22 and the repair service proposal information illustrated in Fig. 23, but may also display on the screen the failure probability and failure location included in the inference results, graph the transition of the time-series data of the IoT information of the air conditioning equipment 200 and display it on the screen, and display customer information and model name information of the air conditioning equipment 200 on the screen.

[0084] In step S129, the response information input unit 113 accepts an input to select one of the response proposals displayed by the display unit 112 through a response history storage process illustrated in Fig. 19 , and stores the selected response proposal in the response history database 103. Specifically, for example, if a unit replacement repair service derived by the repair service proposal generation unit 111 is selected, the response information input unit 113 generates response history information including the serial number of the air conditioning equipment 200 and information that the selected response proposal is unit replacement, and stores the information in the response history database 103. When the new response history information is stored in the response history database 103, this response history information and the time-series data of the IoT information are transmitted to the learning device 106 through an information transmission process illustrated in Fig. 7 , and are used as learning data for the next learning process.

[0085] As described above, the proposal support device 100 outputs failures and response proposals for the monitored air conditioning equipment 200 within a set period using a trained model trained using training data including time-series data on the IoT information of the air conditioning equipment 200, the failure status of the air conditioning equipment 200, and responses, including repairs or renewal sales, performed on the air conditioning equipment 200. This makes it possible to predict possible failures and propose appropriate responses. Furthermore, a user of the proposal support device 100 can refer to the inference results and make proposals, including repairs and renewal sales, to a customer before a failure occurs in the air conditioning equipment 200.

[0086] As long as it does not deviate from the gist of the present disclosure, it is possible to select and discard the configurations given in the above embodiments, or to change them to other configurations as appropriate.

[0087] In the above embodiment, the example has been described in which the monitored device is an air conditioner, but the present invention is not limited to this. The monitored device may also display the operating status of home appliances such as FA (Factory Automation) devices in a production system, an inspection system, or a processing system, or lighting devices.

[0088] In the above embodiment, the inference device 107 may output the failure probability within a plurality of set periods. For example, the failure probability may be output within one month, within three months, and within one year.

[0089] Furthermore, in the above embodiment, the production management information stored in the production management database 108 and the parts management information stored in the parts management database include cost information for renewal sales and repair services, and the renewal sales proposal generation unit 109 and the repair service proposal generation unit 111 may calculate the costs to be charged to the customer by implementing each generated proposal and include this in the proposal information.

[0090] Furthermore, the information stored in the storage unit 14 may be collectively managed by a cloud server on the network, and the proposal support device 100 may access the cloud server as needed to read and write information. In this case, the proposal support device 100 does not need to include the IoT information database 102, the response history database 103, the production management database 108, and the parts management database 110.

[0091] Furthermore, the proposal support device 100 can be realized using a normal computer system, rather than a dedicated device. For example, a program for realizing each function may be stored and distributed on a computer-readable recording medium such as a CD-ROM (Compact Disc Read Only Memory) or a DVD-ROM (Digital Versatile Disc Read Only Memory), and a computer capable of realizing each of the above-described functions may be configured by installing the program on a computer.

[0092] Furthermore, when each function is realized by sharing the functions between an OS (Operating System) and an application, or by cooperation between the OS and the application, only the application may be stored on the recording medium.

[0093] The present disclosure allows various embodiments and modifications without departing from the broad spirit and scope of the present disclosure. Furthermore, the above-described embodiments are intended to explain the present disclosure and do not limit the scope of the present disclosure. In other words, the scope of the present disclosure is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and the meaning of equivalent disclosures are considered to be within the scope of the present disclosure.

[0094] Various aspects of the present disclosure are summarized below as appendices.

[0095] (Supplementary Note 1) A proposal support device comprising: an IoT information acquisition unit that acquires IoT information from a monitored device, including the operating status of the device and environmental information that indicates the environment of the location where the device is installed; a learning unit that learns the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history that indicates whether or not the device has been repaired or updated; and an inference unit that infers failures that may occur within a set period of time for the device that output the IoT information, based on the time series including the IoT information newly acquired by the IoT information acquisition unit, and outputs a proposal that includes at least one of a repair service for the device and a renewal sale to update the device.

[0096] (Supplementary Note 2) The proposal support device described in Supplementary Note 1, wherein the learning unit comprises: a learning information acquisition unit that acquires the time series of the IoT information, the failure status of the device, and the response history implemented on the device; and a learning model generation unit that generates a trained model indicating the relationship between the time series, the failure status, and the response history by machine learning using learning information based on the time series, the failure status, and the response history acquired by the learning information acquisition unit.

[0097] (Supplementary Note 3) The proposal support device described in Supplementary Note 2 includes: an inference information generation unit that acquires the IoT information newly acquired by the IoT information acquisition unit and the past IoT information of a device that has output the newly acquired IoT information from an IoT information database that accumulates the IoT information, and generates inference information including the time series; and an inference result generation unit that inputs the inference information generated by the inference information generation unit into the trained model, and generates an inference result including the failure probability within the period, the failure location, and the proposal.

[0098] (Supplementary Note 4) The proposal support device described in any one of Supplementary Notes 1 to 3 further comprises a renewal sales proposal generation unit that, when the inference unit detects that the renewal sales proposal has been output, refers to a production management database that stores production plans for each model name of the equipment, and generates renewal sales proposal information including the model name of the equipment to be processed and the shipping date of at least one model name of an alternative model name.

[0099] (Supplementary Note 5) The proposal support device described in any one of Supplementary Notes 1 to 3 further comprises a repair service proposal generation unit that, when the inference unit detects that the repair service proposal has been output, generates repair service proposal information including a time when the repair service included in the proposal can be implemented by referring to a parts management database that stores a parts production or procurement plan for each repair service.

[0100] (Supplementary Note 6) The proposal support device according to any one of Supplementary Notes 1 to 5, further comprising: a response information input unit that accepts a selection of any of the proposals output by the inference unit, generates response history information including the selected proposal, identification information that uniquely identifies the device, and the failure status of the device, and stores the generated response history information in a response history database that stores response histories for the device.

[0101] (Supplementary Note 7) A proposal support method in which a computer executes the steps of: acquiring IoT information from a monitored device, the IoT information including the operating status of the device and environmental information indicating the environment of the location where the device is installed; learning the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history indicating whether or not the device has been repaired or updated; and inferring, based on the time series including the newly acquired IoT information, failures that may occur within a set period of time for the device that output the IoT information, and outputting a proposal including at least one of a repair service for the device and a renewal sale to update the device.

[0102] (Supplementary Note 8) A program that causes a computer to execute the following processes: a process of acquiring IoT information from a monitored device, the IoT information including the operating status of the device and environmental information indicating the environment where the device is installed; a process of learning the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history indicating that either a repair or an update has been performed on the device; and a process of inferring, based on the time series including the newly acquired IoT information, failures that may occur within a set period of time for the device that output the IoT information, and outputting a proposal including at least one of a repair service for the device and a renewal sale to update the device.

[0103] It should be noted that the present disclosure allows various embodiments and modifications without departing from the broad spirit and scope of the present disclosure. Furthermore, the above-described embodiments are intended to explain the disclosure and do not limit the scope of the present disclosure. That is, the scope of the present disclosure is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and within the meaning of the disclosure equivalent thereto are considered to be within the scope of the present disclosure.

[0104] This application is based on Japanese Patent Application No. 2024-066629, filed on April 17, 2024. The entire specification, claims, and drawings of Japanese Patent Application No. 2024-066629 are incorporated herein by reference.

[0105] 100 Proposal support device, 200 Air conditioning equipment, 101 IoT information acquisition unit, 102 IoT information database, 103 Response history database, 104 IoT information output unit, 105 Response history output unit, 106 Learning device, 107 Inference device, 108 Production management database, 109 Renewal sales proposal generation unit, 110 Parts management database, 111 Repair service proposal generation unit, 112 Display unit, 113 Response information input unit, 1061 Data acquisition unit, 1062 Data pre-processing unit, 1063 Learning model generation unit, 1064 Learned model accumulation unit, 1071 Data acquisition unit, 1072 Data pre-processing unit, 1073 Inference unit, 11 CPU, 12 RAM, 13 ROM, 14 Storage unit, 15 Input unit, 16 Display unit, 17 Communication unit, 20 Customer master, 21 Model name master, 22 Calendar master, 99 Internal bus.

Claims

1. A proposal support device comprising: an IoT information acquisition unit that acquires IoT information from a monitored device, including the operating status of the device and environmental information indicating the environment where the device is installed; a learning unit that learns the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history indicating whether or not the device has been repaired or updated; and an inference unit that infers, based on the time series including the IoT information newly acquired by the IoT information acquisition unit, failures that may occur within a set period of time for the device that output the IoT information, and outputs a proposal including at least one of a repair service for the device and a renewal sale to update the device.

2. The proposal support device described in claim 1, wherein the learning unit comprises: a learning information acquisition unit that acquires the time series of the IoT information, the failure status of the device, and the response history implemented on the device; and a learning model generation unit that generates a trained model indicating the relationship between the time series, the failure status, and the response history by machine learning using learning information based on the time series, the failure status, and the response history acquired by the learning information acquisition unit.

3. The proposal support device described in claim 2, wherein the inference unit comprises: an inference information generation unit that acquires the IoT information newly acquired by the IoT information acquisition unit and the past IoT information of the device that output the newly acquired IoT information from an IoT information database that stores the IoT information, and generates inference information including the time series; and an inference result generation unit that inputs the inference information generated by the inference information generation unit into the trained model, and generates an inference result including the failure probability within the period, the failure location, and the proposal.

4. A proposal support device as described in any one of claims 1 to 3, further comprising a renewal sales proposal generation unit that, when the inference unit detects that the renewal sales proposal has been output, references a production management database that stores production plans for each model name of the equipment, and generates renewal sales proposal information including the model name of the equipment to be processed and the shipping date of at least one alternative model name.

5. A proposal support device as described in any one of claims 1 to 3, further comprising a repair service proposal generation unit that, when the inference unit detects that the repair service proposal has been output, references a parts management database that stores parts production or procurement plans for each repair service, and generates repair service proposal information including a time when the repair service included in the proposal can be implemented.

6. A proposal support device as described in any one of claims 1 to 5, further comprising a response information input unit that accepts a selection of any of the proposals output by the inference unit, generates response history information including the selected proposal, identification information that uniquely identifies the device, and the failure status of the device, and stores the generated response history information in a response history database that stores the response history for the device.

7. A proposal support method in which a computer executes the following steps: acquiring IoT information from a monitored device, the IoT information including the operating status of the device and environmental information indicating the environment where the device is installed; learning the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history indicating whether or not the device has been repaired or updated; and inferring, based on the time series including the newly acquired IoT information, failures that may occur within a set period of time for the device that output the IoT information, and outputting a proposal including at least one of a repair service for the device and a renewal sale to update the device.

8. A program that causes a computer to execute the following processes: a process of acquiring IoT information from a monitored device, including the operating status of the device and environmental information indicating the environment where the device is installed; a process of learning the relationship between a time series of the IoT information, a failure status including whether or not the device is malfunctioning and the details of the failure, and a response history indicating whether or not the device has been repaired or updated; and a process of inferring, based on the time series including the newly acquired IoT information, failures that may occur within a set period of time for the device that output the IoT information, and outputting a proposal including at least one of a repair service for the device and a renewal sale to update the device.

Images