Information processing device, information processing method, and program

Abstract

An information processing device, an information processing method, and a program are provided that support multifaceted failure analysis. [Solution] The information processing device 10 has a control unit 200 which is provided with a first acquisition means 211 which acquires component failure data, a storage means 220 which stores the failure data, a second acquisition means 212 which acquires an instruction to create an analysis result of the failure data, the instruction including a predetermined first factor related to the failure, an extraction means 232 which extracts information related to the first factor from the failure data stored in the storage unit 100, an analysis execution means 233 which receives information related to the predetermined factor related to the failure as input and generates information related to the factor and the timing of the failure based on information related to the first factor using a mathematical model which outputs information related to the factor and the timing of the failure, an output data generation means 241 which generates output data which outputs information related to the first factor and the timing of the failure, and an output means 250 which outputs an analysis result of the failure data including the generated output data.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] 2. Description of the Related Art There has been a demand for a system that supports the analysis of failures in parts of vehicles and the like.

[0003] For example, Patent Document 1 discloses a system that predicts vehicle failures based on live data relating to the operating status of devices that make up the vehicle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2023-143537 Summary of the Invention [Problem to be solved by the invention]

[0005] However, conventional failure detection technologies, including the failure prediction system described in Patent Document 1, attempt to detect failures in each product by using a specific diagnostic device or program for a specific product. In other words, conventional failure detection is limited to failure analysis of individual products or parts, and does not perform multifaceted failure analysis of parts.

[0006] The present invention aims to support multifaceted failure analysis. [Means for solving the problem]

[0007] an extraction means for extracting information relating to the first factor from the failure data stored in the storage unit based on the creation instruction; an analysis execution means for generating information relating to the first factor and the timing of the failure based on the extracted information relating to the first factor using a mathematical model that performs statistical processing, the mathematical model receiving information relating to the predetermined factor related to the failure as input and outputting information relating to the factor and the timing of the failure; an output data generation means for generating output data that outputs information relating to the generated first factor and the timing of the failure; and an output means for outputting an analysis result of the failure data including the generated output data.

[0008] An information processing method according to one aspect of the present invention includes a first acquisition step in which a computer acquires failure data related to component failures; a storage step in which the acquired failure data is stored in a storage unit; a second acquisition step in which a user inputs an instruction to create an analysis result of the failure data, the instruction including output data for outputting information related to the timing of the component failure, the instruction including a predetermined first factor related to the failure; an extraction step in which, based on the instruction, information related to the first factor is extracted from the failure data stored in the storage unit; an analysis execution step in which, based on the extracted information related to the predetermined first factor, information is generated related to the first factor and the timing of the failure using a mathematical model that performs statistical processing and receives information related to the predetermined factor related to the failure and outputs information related to the factor and the timing of the failure; an output data generation step in which output data for outputting the generated information related to the first factor and the timing of the failure; and an output step in which the analysis result of the failure data including the generated output data is output.

[0009] a first acquisition step of acquiring failure data related to component failures; a storage step of storing the acquired failure data in a storage unit; a second acquisition step of acquiring instructions for creating an analysis result of the failure data, the instruction being input by a user and including output data for outputting information related to the timing of the component failure, the instruction including a predetermined first factor related to the failure; an extraction step of extracting information related to the first factor from the failure data stored in the storage unit based on the creation instruction; an analysis execution step of generating information related to the first factor and the timing of the failure based on the extracted information related to the predetermined first factor using a mathematical model that performs statistical processing and receives information related to the predetermined factor related to the failure and outputs information related to the factor and the timing of the failure; an output data generation step of generating output data for outputting the generated information related to the first factor and the timing of the failure; and an output step of outputting the analysis result of the failure data including the generated output data. [Effects of the Invention]

[0010] According to the present invention, multifaceted failure analysis can be supported. [Brief explanation of the drawings]

[0011] [Figure 1] 1 shows an example of the configuration of an information processing system. [Figure 2] FIG. 1 is a diagram illustrating an example of the operation of an information processing system. [Figure 3] FIG. 1 is a diagram illustrating an example of the operation of an information processing system. [Figure 4] 1 shows an example of the hardware configuration of an information processing device. [Figure 5] 1 shows an example of a functional configuration of an information processing device. [Figure 6] 10 is an example of failure data. [Figure 7] 10 shows an example of an analysis result creation instruction screen. [Figure 8]10 shows an example of an analysis result display screen. [Figure 9] 10 shows an example of an analysis result display screen. [Figure 10] 10 shows an example of an analysis result display screen. [Figure 11] 10 shows an example of an analysis result display screen. [Figure 12] 10 shows an example of an analysis result display screen. [Figure 13] 10 shows an example of an analysis result creation instruction screen. [Figure 14] 10 shows an example of an analysis result display screen. [Figure 15] 10 shows an example of an analysis result creation instruction screen. [Figure 16] FIG. 10 is a sequence diagram illustrating an example of the operation of the information processing system. DETAILED DESCRIPTION OF THE INVENTION

[0012] 1. Information processing system configuration 1 illustrates an example of a configuration of an information processing system 1 according to one embodiment of the present disclosure. The information processing system 1 includes an information processing device 10 and a user device 20. The devices constituting the information processing system 1 are connected to each other via a communication network N so as to be able to communicate with each other.

[0013] The communication network N is a wired or wireless network. Examples include the Internet, a LAN (Local Area Network), a dedicated line, a telephone line, an in-house network, a mobile communication network, Bluetooth (registered trademark), WiFi (Wireless Fidelity), other communication lines, and combinations thereof. Each component of the information processing system 1 may be located in the same facility or in different facilities.

[0014] The information processing device 10 has a functional configuration described below and is a device that executes various information processes based on various information input from the user device 20. The information processing device 10 is not particularly limited, and examples thereof include a desktop, a laptop, and other dedicated or general-purpose computers. The information processing device 10 may be configured as a single computer, or may be configured as multiple computers on a communication network N.

[0015] The user device 20a is an information processing device used by a user of a product including a part that is the target of failure analysis (hereinafter, sometimes referred to as a "product user.") The product user, for example, operates the user device 20a to notify a manufacturer or a retailer of a failure.

[0016] This disclosure describes a case where the target of failure analysis is a vehicle part, but the scope of application of this disclosure is not limited to this. For example, the disclosure can also be applied to home appliances such as refrigerators and air conditioners, precision instruments such as smartphones and printers, industrial equipment such as generators and elevators, medical equipment such as ultrasound diagnostic devices and X-ray devices, transportation-related equipment such as railway systems and power transmission facilities, or infrastructure-related equipment. Furthermore, while this disclosure describes a case where a part fails, the contents of this disclosure can also be applied to a case where a product consisting of multiple parts fails.

[0017] The user device 20b is an information processing device used by a user of a dealer of a product containing a part that is the subject of failure analysis (hereinafter, sometimes referred to as a "dealer user"). The dealer user, for example, operates the user device 20b to check information about the failure notified by the product user. The dealer user also, for example, operates the user device 20b to transmit failure data to an insurance company.

[0018] A retailer user may provide a warranty service for malfunctions in cooperation with a warranty company. The warranty company may be responsible for system design, operation, and repair cost payment of the warranty service provided in cooperation with the retailer. The warranty company may also conclude an insurance contract with a non-life insurance company and pay the insurance premium. Specifically, for example, the retailer pays the insurance premium to the insurance company. Furthermore, when a product user who is a warranty service subscriber requests a repair (notifies of a malfunction), the retailer either performs the repair, or the retailer and the warranty company work together to request the repair from the manufacturer or an affiliated company that provides repair services. The retailer, manufacturer, or affiliated company performs the repair, and once the repair is complete, the product is delivered to the product user. The retailer and the warranty company then work together to pay the repair costs to the manufacturer or affiliated company, and receive the insurance money from the insurance company.

[0019] The user device 20c is an information processing device used by a user of the manufacturer of the part that is the subject of failure analysis (hereinafter, sometimes referred to as a "manufacturer user"). The manufacturer user, for example, operates the user device 20c to acquire failure data transmitted from a dealer user. In addition, the dealer user, for example, operates the user device 20c to transmit information regarding payment of repair costs to the dealer. FIG. 2 is a diagram illustrating an example of the flow of information when a dealer and a manufacturer cooperate to compensate for damages caused by a breakdown. A product user notifies a dealer of a breakdown (S10), and after repairs are performed at the dealer (S11), the repaired part (or the vehicle including the repaired part) is delivered to the product user (S12). The dealer user then transmits breakdown data related to the breakdown to the manufacturer and bills the manufacturer for repair costs (S13), and the manufacturer user receives the breakdown data (S14). The dealer may compile the breakdown data related to the notified breakdown at predetermined intervals, such as once a month, and transmit it to the manufacturer. The manufacturer considers compensation for damages caused by the breakdown (payment of repair costs) and transmits information regarding payment of repair costs to the dealer (S15). Compensation for damages caused by a breakdown involving the information flow shown in FIG. 2 is sometimes called a manufacturer's warranty, as it is a system in which the manufacturer guarantees compensation for damages.

[0020] The user device 20d is an information processing device used by a user (hereinafter, sometimes referred to as an "insurance company user") who provides insurance services related to damages and the like when a breakdown occurs. The insurance company user, for example, operates the user device 20d to acquire breakdown data transmitted from a dealer user. The insurance company user also, for example, operates the user device 20d to transmit information regarding insurance payment to the dealer. FIG. 3 is a diagram illustrating an example of the flow of information when a dealer and an insurance company cooperate to compensate for damages caused by a breakdown. A product user notifies a dealer of a breakdown (S20). After the dealer performs repairs (S21), the repaired part (or the vehicle including the repaired part) is delivered to the product user (S22). The dealer user then transmits breakdown data related to the breakdown to the insurance company and claims insurance (S23). For example, the breakdown data may be transmitted to the insurance company because the warranty period of the manufacturer's warranty shown in FIG. 2 has expired. The dealer may also compile the breakdown data related to the notified breakdown at predetermined intervals, such as every month, and transmit it to the insurance company. The insurance company user then receives the breakdown data (S24). The insurance company considers compensation for damages caused by the breakdown (payment of insurance claims), and transmits information regarding the payment of insurance claims to the dealer (S25). Compensation for damages caused by a breakdown involving the information flow shown in FIG. 3 is sometimes called an extended warranty, as it ensures compensation for damages for a certain period of time even after the manufacturer's warranty period has expired.

[0021] In this disclosure, repair costs and insurance payments incurred when compensating for damages caused by a malfunction may be referred to as compensation.

[0022] The user device 20a, the user device 20b, the user device 20c, and the user device 20d are not particularly limited, and may be, for example, a mobile phone, a smartphone, a desktop, a laptop, a tablet terminal device, or the like.

[0023] In this disclosure, when there is no particular distinction between product users, dealer users, manufacturer users, and insurance company users, they will be referred to as users. Furthermore, when there is no particular distinction between user device 20a, user device 20b, user device 20c, and user device 20d, they will be referred to as user device 20.

[0024] In the present disclosure, the information processing device 10 may use a combination of failure data transmitted from a dealer to an insurance company and failure data transmitted from a manufacturer to an insurance company. This enables comprehensive failure analysis. The information processing device 10 may also execute various processes of the present disclosure based on data related to insurance claims for failures, among the failure data. This also enables analysis focusing on abnormalities and frauds related to insurance claims for failures.

[0025] 2. Hardware Configuration 4 illustrates an example of a hardware configuration of the information processing device 10. The information processing device 10 includes a processor 11, a storage device 12, a communication IF (Interface) 13, an input device 14, and an output device 15.

[0026] The processor 11 is a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.

[0027] The storage device 12 is a memory, a hard disk drive (HDD), and / or a solid state drive (SSD), etc. The memory is, for example, a random access memory (RAM) and / or a read only memory (ROM).

[0028] The communication IF 13 is a device that performs data communication with other devices via a communication network N in a wired or wireless manner.

[0029] The input device 14 accepts input of information, and may be, for example, a keyboard, a touch panel, a mouse, a camera, and / or a microphone.

[0030] The output device 15 outputs information, and is, for example, a display, a touch panel, and / or a speaker.

[0031] The hardware configuration of the user device 20 can be the same as that of the information processing device 10, and therefore a description thereof will be omitted.

[0032] 3. Functional Configuration 5 illustrates an example of a functional configuration of the information processing device 10. The information processing device 10 has a storage unit 100 and a control unit 200. The storage unit 100 can be realized by a storage unit 12 included in the information processing device 10. The control unit 200 can be realized by a processor 11 included in the information processing device 10 executing a program stored in the storage unit 12. The program can be stored in a storage medium. The storage medium storing the program may be a non-transitory computer-readable medium. The non-transitory storage medium may be, for example, a Universal Serial Bus (USB) memory or a compact disc read-only memory (CD-ROM).

[0033] (Storage part) The storage unit 100 stores various types of information required for the information processing device 10 to operate.

[0034] The storage unit 100 can store, for example, failure data acquired by a first acquisition unit 211 (described later). The failure data is data relating to failures of components.

[0035] FIG. 6 illustrates an example of an overview of failure data. In the example of FIG. 6, the following are stored: a failure ID (a10) that identifies the failure occurrence incident; the part that failed in the incident (a11); factors 1 to 4 (a12 to a15) that are (potentially) involved in the failure in the incident; the date and time of the occurrence of the incident (a16); and other support information (a17) related to the incident. In the example of FIG. 6, the region is stored as factor 1, the manufacturer as factor 2, the dealer as factor 3, and the vehicle model as factor 4. Examples of other support information include the cumulative mileage of the vehicle at the time the incident occurred, the contract period for damage compensation, the details of the contract for damage compensation, the amount of insurance and / or repair costs claimed in the incident, and insurance and / or repair cost claim data (failure claim data) related to the failure, such as the source and destination of the claim for insurance and / or repair costs in the incident.

[0036] (Control unit) The control unit 200 may include an acquiring unit 210 , a storing unit 220 , an analyzing unit 230 , a generating unit 240 , and an output unit 250 .

[0037] (Acquisition means: 1st acquisition means) The acquisition means 210 acquires various types of information during the operation of the information processing system 1. In the present disclosure, acquiring information may include accepting input of information, receiving information from other devices via the communication network N, and reading out information stored in the storage unit 100.

[0038] The acquiring means 210 may include a first acquiring means 211. The first acquiring means 211 acquires fault data from the user device 20.

[0039] (memory means) The storage means 220 stores the acquired failure data in the storage unit 100 .

[0040] (Acquisition means: 2nd acquisition means) The acquiring means 210 may include a second acquiring means 212. The second acquiring means 212 acquires from the user device 20 an instruction to create an analysis result obtained by analyzing the failure data (analysis result creation instruction), which is input by the user.

[0041] The instruction to create the analysis result may include predetermined factors related to the failure specified by the user. The factors related to the failure may be specified by the type and content of the factor. Examples of the type of factor include the part, region, season, manufacturer, dealer, or vehicle model related to the failure. Examples of the content of the factor include, for example, a blinker, headlight, or air conditioner if the type is a faulty part. Examples of the content of the factor include Japan, Tokyo, Chiba, Niigata, Hokkaido, etc. if the type is a region. Examples of the content of the factor include information identifying the manufacturer, such as manufacturer M21 or manufacturer M23.

[0042] By issuing an instruction to create an analysis result that includes the specification of factors related to the failure, an analysis is performed (analysis execution means, described later) on the failure data of cases where the factors satisfy the specified type and content (extraction means, described later), output data is generated (output data generation means, described later), and an analysis result including the output data can be created (analysis result creation means, described later).

[0043] Fig. 7 is an example of a screen (analysis result creation instruction screen) when a user inputs an instruction to create an analysis result. The screen S1 shown in Fig. 7 displays an object d100 for specifying a predetermined area as a factor related to the failure, an object d101 for displaying a map for specifying a predetermined area as a factor related to the failure, and an object d102 for specifying a predetermined vehicle model as a factor related to the failure.

[0044] Fig. 13 is an example of another screen (analysis result creation instruction screen) when the user inputs an instruction to create an analysis result. Screen S2 shown in Fig. 13 displays an object d700 for specifying a specific mechanism as a cause of the failure, and an object d701 for specifying a specific part as a cause of the failure.

[0045] Fig. 15 is an example of another screen (analysis result creation instruction screen) when the user inputs an instruction to create an analysis result. Screen S3 shown in Fig. 15 displays an object d900 for inputting in text the vehicle model to be specified as a cause of the failure and searching, and an object d901 for selecting a desired vehicle model as a cause of the failure from the searched vehicle models.

[0046] The user can specify the cause of the failure (input conditions related to the cause) by operating on the objects shown in Figures 7, 13, and 15. For example, in the map object d101 shown in Figure 7, the user can select the Kanto region to specify the cause "area (type) is Kanto region (content)."

[0047] When acquiring the analysis result creation instruction by displaying the analysis result creation instruction screen on user device 20, second acquisition means 212 may display an object for the user to input an instruction on the analysis result creation instruction screen. In response to input of a selection operation for the object, second acquisition means 212 may acquire the analysis result creation instruction.

[0048] (Analysis method: Model building method) The analysis means 230 performs an analysis of the failure data based on the acquired failure data and the analysis result creation instruction. The analysis means 230 may include a model construction means 231. The model construction means 231 constructs a mathematical model M based on the acquired failure data.

[0049] The mathematical model M is a mathematical model that executes statistical processing, for example, with factors related to component failure and failure data as input, and outputs information related to the timing at which a failure is predicted.

[0050] The mathematical model M may be constructed by utilizing breakdown claim data, creating and comparing multiple algorithms, and then constructing an algorithm with features that are correlated with breakdown claims.

[0051] The mathematical model M may be configured to calculate the N-year survival probability and failure prediction value (for example, information on the timing of failure) for each component from the failure data.

[0052] The mathematical model M may use statistical analysis to compare predetermined failure data with other failure data and perform scoring and tagging to detect anomalies. For example, failure claim data from a predetermined dealer as the insurance claim source may be compared with failure claim data from other dealers in the same region, and failure data showing a statistically significant difference may be assigned a predetermined score or a label indicating an anomaly, which may serve as an index of reliability. The mathematical model M may enable scoring and graphing of abnormal values ​​based on multiple failure-related factors. The mathematical model M may also assist in verifying abnormal values ​​and estimating the cause of a failure.

[0053] The mathematical model M is not particularly limited, but may be, for example, any AI model, Kaplan-Meier estimator, linear regression, random forest, support vector machine (SVR), decision tree regression, neural network, gradient boosting tree (XGBoost, LightGBM), k-nearest neighbor method (KNN), polynomial regression, ridge regression, lasso regression, elastic net, Gaussian process regression, etc.

[0054] By using the mathematical model M described above, for example, it may be possible to build an AI algorithm tailored to each manufacturer's failure data, and to individually identify correlated features. Furthermore, statistical analysis may enable estimation of the survival probability N years from now (N-year survival probability) and the failure prediction value. In this disclosure, the estimated survival probability may be referred to as the estimated survival probability. Furthermore, by examining abnormal values ​​in claims from multiple angles, it may be possible to narrow the scope of investigation into factors that are difficult to identify through statistical analysis, thereby assisting in efficient cause investigation.

[0055] For example, in the case of automobiles, big data can be used to determine that mileage affects failure rates, and this information can be used to build the algorithm for mathematical model M. Furthermore, it can also be used to calculate the Y% failure rate of shock absorbers after N years, or the Z% failure rate of air conditioners after N years. Furthermore, for example, abnormal values ​​detected based on a comparison of the same part between vehicle models can provide information for examining whether there are design issues with the part in that vehicle model or whether a recall is necessary. Furthermore, for example, abnormal values ​​detected based on a comparison of the same part between regions can provide information for examining regional factors. Specifically, for example, if the survival probability in Hokkaido is low, it can provide information for examining whether temperature differences or hot and cold temperatures are the cause. Furthermore, for example, abnormal values ​​detected based on a comparison of dealerships in the same region can provide information for examining whether there are inconsistencies in billing standards or whether there is a possibility of fraudulent billing.

[0056] For example, in the case of refrigerators, big data can be used to determine that the number of years of use and the average temperature in the region of use affect the failure rate, and this information can be used to build the algorithm for mathematical model M. The failure rate (%) for each refrigerator model after N years can be calculated. Furthermore, for example, anomalies detected based on a comparison of the same part's use between regions can provide information for identifying failure factors unique to a specific region. Specifically, this can provide information for examining the possibility that differences in voltage or differences in usage habits in a specific region overseas result in refrigerators being packed with large amounts of food, which places a greater strain on part X. Furthermore, for example, anomalies detected based on a comparison of the same part's use between retailers can provide information for examining failure factors unique to each retailer, such as the possibility that retailer X has quality control issues, such as product storage conditions, which may be affecting the failure rate. Furthermore, if the failure rate of a particular part is particularly high, for example, this can provide information for examining whether the part is defective and whether a recall is necessary.

[0057] (Analysis means: Extraction means) The analysis means 230 may include an extraction means 232. Based on an analysis result creation instruction, the extraction means 232 extracts information related to a predetermined cause from the failure data stored in the storage unit 100. For example, the extraction means 232 extracts failure cases whose causes satisfy the types and contents of causes related to the failure specified by the user and included in the analysis result creation instruction.

[0058] Specifically, for example, if the user specifies the factor "region (type) is Kanto region (content)", the extraction means 232 extracts failure cases related to a factor that satisfies the content that the "region" type is "Kanto region" from the failure data stored in the storage unit 100. In the case of the failure data shown in Fig. 6, the extraction means 232 extracts a case in which factor 1 of the "region" type is "Tokyo" (case with failure ID 001) and a case in which factor 1 of the "region" type is "Chiba" (case with failure ID 002).

[0059] (Analysis method: Analysis execution method) The analysis means 230 may include an analysis execution means 233. The analysis execution means 233 generates information (analysis information) obtained by performing a statistical analysis based on information relating to predetermined factors extracted from the failure data.

[0060] The analysis execution means 233 may generate information relating to the cause and the timing of the failure, for example. The analysis execution means 233 may generate information relating to the cause and the timing of the failure using a mathematical model M.

[0061] The information relating to the specified cause and the timing of the failure may include, for example, information indicating the tendency of failures to occur in cases where the failure satisfies the type and content specified by the user for the specified cause.

[0062] Examples of the tendency of failures include the probability of failure, the number of occurrences of failures, the change in the number of occurrences of failures, and the interval between occurrences of failures. The tendency of failures may be a tendency analyzed according to the mileage of the vehicle in each case. For example, it may be a change in the tendency of failure according to the mileage, the survival probability, etc. When the subject of failure analysis is something other than a vehicle, the mileage may be the period of use or the amount of use.

[0063] For example, suppose the first cause specified by the user is "region (type) is Kanto region (content)." In this case, in the case of the failure data in Fig. 6, information indicating the occurrence tendency of failures in the case of failure ID 001 and the case of failure ID 002 that satisfy the type and content of the first cause becomes information related to the first cause and the timing of the failure.

[0064] The same applies when the user specifies two or more types of factors. For example, suppose the factors specified by the user are "region (type) is the Kanto region (content)" as the first factor and "manufacturer (type) is a manufacturer identified by M21 (content)" as the second factor. In this case, in the case of the failure data in FIG. 6, information indicating the occurrence tendency of failures in the case of failure ID 001 and the case of failure ID 002 that satisfy the types and contents of the first and second factors becomes information regarding the first and second factors and the timing of the failures. The same applies when the user specifies two or more types of factors, or when there are two or more types and two or more contents.

[0065] The analysis execution means 233 may generate information relating to a predetermined cause and insurance claims based on information relating to the cause extracted from the failure data. The information relating to a predetermined cause and insurance claims may include, for example, information relating to insurance claims for breakdown cases that satisfy the type and content specified by the user for the predetermined cause. The information relating to insurance claims may be, for example, information relating to insurance claims such as the number of insurance claims, the percentage of insurance claims, and the amount of insurance claims. For example, suppose the first cause specified by the user is "the region (type) is the Kanto region (content)." In this case, based on the support information for each breakdown case that satisfies the type and content of the first cause, information relating to insurance claims such as the number of insurance claims (total), the percentage of insurance claims ([number of insurance claims] / [number of held contracts (number of contracted contracts)]), and the frequency of insurance claims ([number of insurance claims] / [number of contracted vehicles]) becomes the information relating to the first cause and insurance claims. In one aspect, the information regarding the insurance claim may be information regarding the repair cost claim, or may be information regarding the insurance claim and the repair cost claim.

[0066] Similarly, in parallel with generating the analysis information, the analysis execution means 233 may further generate information (supplementary information) regarding other factors not specified by the user in a breakdown case related to a factor that satisfies the type and content specified by the user, information regarding the timing of the breakdown, or information regarding insurance claims. For example, assume that the first factor specified by the user is "region (type) is the Kanto region (content)." In this case, in the case of the breakdown data of FIG. 6, the analysis execution means 233 generates information regarding insurance claims for the case of breakdown ID 001 and the case of breakdown ID 002 that satisfy the type and content of the first factor. In parallel with this, the analysis execution means 233 further generates information regarding insurance claims for each type of vehicle, which is a type of other factor not specified by the user, for example. Specifically, for example, the analysis execution means 233 sets a first supplementary factor of "the vehicle model (type) is vehicle model A (content)", and calculates (generates) information related to insurance claims for cases that satisfy the type and content of the first supplementary factor among the cases of failure ID 001 and failure ID 002 that satisfy the type and content of the first supplementary factor as supplementary information. The supplementary information may be passed to the output data generation means 241, which will be described later, together with the analysis information. The supplementary information may be stored in the storage unit 100 so as to be able to be referenced.

[0067] The supplemental information is not limited to targeting failure cases related to factors that satisfy the type and content specified by the user. For example, the analysis execution means 233 may generate supplemental information regardless of the progress of the analysis information generation process based on the analysis result creation instruction. For example, without being triggered by a user's instruction to create an analysis result, the analysis execution means 233 may automatically execute an analysis of failure data for a predetermined period (e.g., one year up to 2025) stored in the storage unit 100, analyze trends in failure cases that occurred in 2025, and generate the results as supplemental information. Also, without being triggered by a user's instruction to create an analysis result, the analysis execution means 233 may automatically execute an analysis of failure data in Japan stored in the storage unit 100, analyze trends in failure cases that occurred in 2025, and generate the results as supplemental information. Such analysis processing that is executed regardless of a user's instruction to create an analysis result may be executed automatically based on user attributes stored in association with the user account used to display the analysis result creation instruction screen. Alternatively, the analysis execution means 233 may be executed automatically in response to the display of the analysis result creation instruction screen. The supplemental information thus generated may be displayed on the analysis result generation instruction screen.

[0068] The analysis execution means 233 may compare predetermined failure data with other failure data through statistical analysis and perform scoring and tagging to detect abnormalities. For example, failure data for which a predetermined dealer is the source of the insurance claim may be compared with failure data for which other dealers in the same region are the source of the insurance claim, and a predetermined score or tag may be assigned to failure data for which a statistically significant difference is found, and this may be used as an index of reliability.

[0069] (Generation means: output data generation means) The generating means 240 may include an output data generating means 241. The output data generating means 241 generates output data for outputting the information generated by the analysis executing means 233. A known algorithm may be used to generate the output data.

[0070] The output data elements may include, for example, graphs, tables, diagrams, illustrations, and the like, as well as data (infographics) that may be created by combining these elements, audio, music, and the like. Infographics may also include support information. The output data may also be, for example, visual data that visually represents the information generated by the analysis execution means 233.

[0071] The output data generating means 241 generates output data that outputs information relating to, for example, a predetermined cause and a time of failure, or information relating to, for example, a predetermined cause and an insurance claim.

[0072] In the example of FIG. 8, when the first factor is "area (type) is Kanto region (content)", output data d201 that visually represents the analysis information generated by the analysis execution means 233 is displayed.

[0073] In addition, the output data generation means 241 may further generate output data that visually represents information regarding a specified supplementary factor and the timing of the failure, or information regarding a specified supplementary factor and an insurance claim, based on the supplementary information received from the analysis execution means 233.

[0074] The example of FIG. 8 is an example of an analysis result display screen when the user specifies "region (type) is Kanto region (content)" as the first factor on the analysis result creation instruction screen shown in FIG. 7. Here, the first supplementary factor set by the analysis execution means 233 when generating supplementary information is "vehicle model (type) is model A (content)," the second supplementary factor is "vehicle model (type) is model B (content)," and the third supplementary factor is "vehicle model (type) is model C (content)." In the example of FIG. 8, output data of information d203i regarding insurance money for cases that satisfy the type and content of the first supplementary factor among cases of breakdowns that satisfy the type and content of the first supplementary factor specified by the user is displayed. Similarly, output data of information d203ii regarding insurance money for cases that satisfy the type and content of the second supplementary factor among cases of breakdowns that satisfy the type and content of the first factor specified by the user is displayed. Similarly, output data d203iii regarding insurance money in cases where the type and content of the third supplementary factor are satisfied among cases of breakdowns that satisfy the type and content of the first factor specified by the user is displayed.

[0075] (Generation means: Analysis result creation means) The generating means 240 may include an analysis result generating means 242. The analysis result generating means 242 generates the analysis results by the analysis executing means 233. The analysis results may include generated output data.

[0076] (output means) The output means 250 outputs the analysis results of the generated failure data, for example, by causing the user device 20 to display an analysis result display screen that displays the analysis results.

[0077] The analysis result display screen may include output data generated by the output data generating means 241.

[0078] When the output means 250 displays the analysis result display screen on the user device 20, it may display an object (such as an icon) on the analysis result display screen for the user to input an instruction to create an additional analysis result. In response to input of a selection operation for the object, the second acquisition means 212 may acquire the instruction to create an additional analysis result. The instruction to create an additional analysis result may include a change such as adding or deleting a designation of a factor related to a failure. The instruction to create an additional analysis result may also be an instruction to regenerate output data. After the factors designated by the user are changed, the analysis means 230 may perform the analysis, and the generation means 240 may generate output data and create new analysis results in the same manner as before the change.

[0079] 4.Screen display Examples of screen displays by the information processing system 1 will be described with reference to FIGS.

[0080] 7 is an example of an analysis result creation instruction screen (screen S1). Screen S1 displays an object d100 for the user to specify a specific region as a factor related to the failure, an object d101 that displays a map for specifying a specific region as a factor related to the failure, and an object d102 for specifying a specific vehicle model as a factor related to the failure. Screen S1 also displays supplementary information d103.

[0081] FIG. 8 shows an example of an analysis result display screen (screen A1). Screen A1 is an example of a screen displayed when the user specifies "region (type) is Kanto region (content)" as the first factor on screen S1. Screen A1 displays an object d200 for the user to change the region as a breakdown-related factor, and output data d201 of analysis information related to the first factor. Screen A1 also displays output data d202 of supplemental information generated without a user instruction as a trigger (domestic insurance claim ratio: 6%) and analysis information generated with a user instruction as a trigger (insurance claim ratio in the Kanto region: 7%). Screen A1 also displays output data d203i, d203ii, and d203iii of supplemental information when a vehicle model not specified by the user is set as a supplemental factor. Screen A1 may be configured to allow the user to select a dealer ID (d207). To enable the user to select a dealer ID on screen A1, a link for specifying another factor may be embedded in the output data d201 of the analysis information. As shown on screen A1, the supplementary information may be displayed in a ranking format.

[0082] FIG. 9 is an example of an analysis result display screen (screen A2). Screen A2 is an example of a screen displayed when the user selects a dealer ID on screen A1. When transitioning from screen A1 to screen A2, supplemental information stored in storage unit 100 may be read out, or new analysis information may be generated. When the user selects "dealer ID 14601" on screen A1, a second factor, "dealer (type) is the dealer identified by dealer ID 14601 (content)," may be newly specified. Screen A2 displays information d300 about the reliability of each part in failure data that satisfies the first and second factors, and information d301 about insurance money for the failure data. Screen A2 may also be configured to allow the user to select a part (d307).

[0083] FIG. 10 is an example of an analysis result display screen (screen A3). Screen A3 is an example of a screen that is displayed when the user selects air conditioning condenser C1 on screen A2. When transitioning from screen A2 to screen A3, supplemental information stored in storage unit 100 may be read out, or new analysis information may be generated. When the user selects "air conditioning condenser C1" on screen A2, a third factor, "the part (type) is air conditioning condenser C1 (content)," may be newly specified. Screen A3 displays information d400 related to insurance money for failure data that satisfies the first factor, second factor, and third factor, and information d401 related to the timing of the failure in the failure data.

[0084] 11 is an example of an analysis result display screen (screen B1). Screen B1 is an example of a screen that is displayed when the user specifies on screen S1 that "the region (type) is the Kanto region (content)" as the first factor and "the vehicle model (type) is vehicle model a (content)" as the second factor. Screen B1 displays information d500 related to insurance money for failure data that satisfies the first and second factors, and information d501 related to the timing of the failure in that failure data.

[0085] 12 is an example (B2) of an analysis result display screen. Screen B2 is another example of a screen that is displayed when the user specifies on screen S1 that "the region (type) is the Kanto region (content)" as the first factor and "the vehicle model (type) is vehicle model a (content)" as the second factor. Screen B2 displays information d600 related to insurance money for failure data that satisfies the first and second factors, and information d601 related to the timing of the failure in that failure data.

[0086] 13 is an example of an analysis result creation instruction screen (screen S2). Screen S2 displays an object d700 for the user to specify a specific mechanism as a cause of the failure, and an object d701 for the user to specify a specific part as a cause of the failure.

[0087] 14 is an example of an analysis result display screen (screen C1). Screen C1 is an example of a screen that is displayed when the user specifies "the part (type) is a clock assembly (content)" as the first cause on screen S2. Screen S2 displays an object d800 that allows the user to specify a specific region as a cause of the failure, information d801 regarding insurance money for failure data that satisfies the first cause, and information d802 regarding the time of failure for failure data that satisfies the first cause.

[0088] 15 is an example of an analysis result creation instruction screen (screen S3). Screen S3 displays an object d900 for the user to input and search for a vehicle model to specify as a cause of the failure in text, and an object d901 for selecting a desired vehicle model as a cause of the failure from the searched vehicle models.

[0089] 5.Operation The operation of the information processing system 1 according to one embodiment of the present disclosure will be described with reference to FIG.

[0090] First, the user inputs failure data relating to a component failure by operating the user device 20 (S30).

[0091] The information processing device 10 acquires failure data by the first acquisition means 211 (S31), and stores the acquired failure data in the storage unit 100 by the storage means 220 (S32).

[0092] A user who wishes to analyze failure data inputs an instruction to create an analysis result by operating the user device 20 (S33). Here, it is assumed that the instruction to create an analysis result includes a specification of a first factor, that is, "the area (type) is the Kanto region (content)." For example, the user specifies the first factor, that is, "the area (type) is the Kanto region (content)," by selecting the Kanto region from the map object displayed on the analysis result creation instruction screen displayed on the user device 20.

[0093] The information processing device 10 acquires an analysis result creation instruction by the second acquisition means 212, and extracts information on the first cause from the failure data stored in the storage unit 100 by the extraction means 232 (S33). For example, from the failure data shown in Fig. 6, it extracts a case where cause 1 of the "area" type is "Tokyo" (case with failure ID 001) and a case where cause 1 of the "area" type is "Chiba" (case with failure ID 002), which are failure cases related to a cause that satisfies "the area (type) is the Kanto region (content)".

[0094] The information processing device 10 uses the analysis execution means 233 to perform an analysis on the information related to the extracted first factor, and generates analysis information and supplementary information (S35). Here, the analysis is performed on failure data where "the area (type) is the Kanto region (content)". Furthermore, information related to insurance money (information related to the first factor and insurance money claim (analysis information)) is generated for failure data that satisfies the first factor of "the area (type) is the Kanto region (content)". Furthermore, for vehicle types that are a type of factor not specified by the user, a first supplementary factor of "vehicle type (type) is vehicle type A (content)" is set, and information related to insurance money (supplementary information) is generated for failure data that satisfies the first factor.

[0095] Next, the information processing device 10 causes the output data generation means 241 to generate output data that visually represents the analytical information and supplemental information generated by the analysis execution means 233. In the example of Fig. 8, output data d201 of analytical information and output data d203i of supplemental information are displayed.

[0096] Next, the information processing device 10 causes the analysis result creation means 242 to create an analysis result including the generated output data, and causes the output means 250 to display an analysis result display screen on the user device 20 (S37).

[0097] If the user inputs an instruction to create additional analysis results on the analysis result display screen (S38: YES), the process returns to S34. If no instruction to create additional analysis results is input (S38: NO), the process ends.

[0098] As described above, the present disclosure can support multifaceted failure analysis.

[0099] The present invention is not limited to the above-described embodiment, and can be embodied in various other forms without departing from the spirit of the present invention. Therefore, the above-described embodiment is merely an example in all respects and should not be interpreted as being limiting. For example, the order of the above-described processing steps can be arbitrarily changed or executed in parallel as long as no contradiction occurs in the processing content. [Explanation of symbols]

[0100] 1...information processing system, 10...information processing device, 11...processor, 12...storage device, 13...communication IF, 14...input device, 15...output device, 20...user device, 100...storage unit, 200...control unit, 210...acquisition means, 211...first acquisition means, 212...second acquisition means, 220...storage means, 230...analysis means, 231...model construction means, 232...extraction means, 233...analysis execution means, 234...detection means, 240...generation means, 241...output data generation means, 242...analysis result creation means, 250...output means

Claims

1. a first acquisition means for acquiring failure data relating to a failure of a component; a storage unit for storing the acquired failure data in a storage unit; An instruction to create an analysis result of the failure data, which includes output data input by a user and which outputs information regarding the timing of the failure of the component, and which includes a predetermined first cause of the failure. A second acquisition means for acquiring a creation instruction; an extraction unit that extracts information about the first cause from the failure data stored in the storage unit based on the creation instruction; an analysis execution means for generating information about a first factor and a time of failure based on the extracted information about the first factor using a mathematical model that executes statistical processing and receives information about a predetermined factor related to a failure as an input and outputs information about the factor and a time of failure; an output data generating means for generating output data that outputs information relating to the generated first cause and the time of the failure; an output means for outputting an analysis result display screen that displays an analysis result of failure data including the generated output data, the analysis result display screen including an object for the user to specify a second factor that causes generation of new analysis information different from the analysis information based on the first factor; Equipped with By specifying a second factor through an operation by the user on the analysis result display screen, the extraction means further extracts information about the second cause from the failure data stored in the storage unit based on the second cause; the analysis execution means further generates information relating to the first cause, the second cause, and a time of failure using the mathematical model; the output data generating means further generates output data that outputs information regarding the first cause, the second cause, and a time of failure. Information processing device.

2. the output data includes a graph showing a change in estimated survival probability for each factor related to the failure according to the amount of use or the period of use of the part; The information processing device according to claim 1 .

3. the output means causes a user device to display an analysis result display screen that displays the analysis result, and the analysis result display screen includes an icon that allows the user to instruct regeneration of the output data. The information processing device according to claim 1 .

4. The analysis execution means further generates information relating to the first factor and a claim for compensation; the output data generating means further generates output data that outputs information regarding the first factor and a claim for compensation. The information processing device according to claim 1 .

5. the analysis execution means further generates supplemental information regarding a first supplemental factor, which is different from the first factor and is specified by a user, and a time of failure, based on the information regarding the extracted first factor; the output data generation means generates output data for outputting information relating to the first cause and the time of the failure, the output data including an icon for instructing acquisition of the supplementary information. The information processing device according to claim 1 .

6. the part is a vehicle part, the output data is a graph showing a change in the estimated survival probability of the part according to the mileage of the vehicle for each factor related to the failure; The information processing device according to any one of claims 1 to 4.

7. the part is a vehicle part, the first and second factors relating to the failure are different factors selected from a region in which the part is used, a vehicle model in which the part is used, a dealer of the part, and a manufacturer of the part, The information processing device according to claim 1 .

8. the output data is visual data that visually represents information related to the first cause and the time of the failure. The information processing device according to claim 1 .

9. The computer a first acquisition step of acquiring failure data relating to a failure of a component; a storage step of storing the acquired failure data in a storage unit; An instruction to create an analysis result of the failure data, which includes output data input by a user and which outputs information regarding the timing of the failure of the component, and which includes a predetermined first cause of the failure. a second acquisition step of acquiring a creation instruction; an extraction step of extracting information about the first cause from the failure data stored in the storage unit based on the creation instruction; an analysis execution step of generating information about a first factor and a time of failure based on the extracted information about the first factor by using a mathematical model that executes statistical processing and receives information about a predetermined factor related to a failure and outputs information about the factor and a time of failure; an output data generating step of generating output data that outputs information relating to the generated first cause and the time of the failure; an output step of outputting an analysis result display screen that displays an analysis result of failure data including the generated output data, the analysis result display screen including an object for the user to specify a second factor that causes generation of new analysis information different from the analysis information based on the first factor; Run By specifying a second factor through an operation by the user on the analysis result display screen, In the extracting step, information on the second cause is further extracted from the failure data stored in the storage unit based on the second cause; In the analysis execution step, information relating to the first cause, the second cause, and a time of failure is further generated by the mathematical model; the output data generating step further generates output data that outputs information relating to the first cause, the second cause, and a time of failure; Information processing methods.

10. On the computer, a first acquisition step of acquiring failure data relating to a failure of a component; a storage step of storing the acquired failure data in a storage unit; An instruction to create an analysis result of the failure data, which includes output data input by a user and which outputs information regarding the timing of the failure of the component, and which includes a predetermined first cause of the failure. a second acquisition step of acquiring a creation instruction; an extraction step of extracting information about the first cause from the failure data stored in the storage unit based on the creation instruction; an analysis execution step of generating information about a first factor and a time of failure based on the extracted information about the first factor by using a mathematical model that executes statistical processing and receives information about a predetermined factor related to a failure and outputs information about the factor and a time of failure; an output data generating step of generating output data that outputs information relating to the generated first cause and the time of the failure; an output step of outputting an analysis result display screen that displays an analysis result of failure data including the generated output data, the analysis result display screen including an object for the user to specify a second factor that causes generation of new analysis information different from the analysis information based on the first factor; Execute By specifying a second factor through an operation by the user on the analysis result display screen, In the extracting step, information on the second cause is further extracted from the failure data stored in the storage unit based on the second cause; In the analysis execution step, information relating to the first cause, the second cause, and a time of failure is further generated by the mathematical model; the output data generating step further generates output data that outputs information relating to the first cause, the second cause, and a time of failure; program.

Images