WO2022091399A1 - Dispositif et procédé de calcul de taux de prime d'assurance - Google Patents

Dispositif et procédé de calcul de taux de prime d'assurance Download PDF

Info

Publication number
WO2022091399A1
WO2022091399A1 PCT/JP2020/041003 JP2020041003W WO2022091399A1 WO 2022091399 A1 WO2022091399 A1 WO 2022091399A1 JP 2020041003 W JP2020041003 W JP 2020041003W WO 2022091399 A1 WO2022091399 A1 WO 2022091399A1
Authority
WO
WIPO (PCT)
Prior art keywords
insurance premium
premium rate
rate calculation
insurance
future
Prior art date
Application number
PCT/JP2020/041003
Other languages
English (en)
Japanese (ja)
Inventor
健 森山
隆顕 中村
俊通 栗山
Original Assignee
三菱電機株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to PCT/JP2020/041003 priority Critical patent/WO2022091399A1/fr
Publication of WO2022091399A1 publication Critical patent/WO2022091399A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08—Insurance

Definitions

  • This disclosure relates to an insurance premium rate calculation device and an insurance premium rate calculation method.
  • Patent Document 1 a risk value indicating a risk to a user's behavior or attribute is calculated based on user information indicating the user's behavior or attribute, and the calculated risk value is used to provide financial support for each user.
  • a financial product proposal device that predicts a combination of products (insurance) is described.
  • the user's risk value is a reference value when setting an insurance premium, and is calculated based on the frequency with which the user performs an action belonging to a preset risk category.
  • the insurance premium for the device is generally set based on the condition of the device.
  • the states of the devices will differ from each other depending on the initial state of the devices or the operating environment of the devices, even if the devices have the same model and the same function. For this reason, the state of the equipment in the future varies depending on the individual equipment, and there is a problem that the conventional technique cannot propose an appropriate insurance premium rate for the equipment.
  • the risk value that is the basis for setting the insurance premium is a value calculated based on a preset category, and thus represents the future state of the device. Can't.
  • the present disclosure solves the above-mentioned problems, and aims to obtain an insurance premium rate calculation device and an insurance premium rate calculation method that can propose an insurance premium rate corresponding to a device.
  • the insurance premium rate calculation device includes a data acquisition unit that acquires operation data indicating the operation status of the device and a device state estimation unit that estimates the future state of the device by statistically analyzing the operation data of the device. And a premium rate calculation unit that calculates the premium rate for the device based on the information indicating the future state of the device.
  • the future state of the device is estimated by statistically analyzing the operation data of the device, and the insurance premium rate for the device is calculated based on the information indicating the future state of the device. It is possible to propose a premium rate that has been set.
  • FIG. 4A is a graph showing the correspondence relationship between the remaining life of the device and the insurance premium rate
  • FIG. 4B is a graph showing the correspondence relationship between the failure rate of the device and the insurance premium rate
  • FIG. 5A is a block diagram showing a hardware configuration that realizes the function of the insurance premium rate calculation device according to the first embodiment
  • FIG. 5B is a software that realizes the function of the insurance premium rate calculation device according to the first embodiment. It is a block diagram which shows the hardware configuration to execute.
  • FIG. 1 is a block diagram showing a configuration of the insurance setting support system 1 according to the first embodiment.
  • the insurance setting support system 1 shown in FIG. 1 is a system that supports the setting of insurance for the device 2.
  • the insurance setting support system 1 proposes an insurance plan for the device 2 based on the insurance premium rate calculated using the operation data of the device 2.
  • the device 2 is a device capable of continuously acquiring operation data, for example, an air conditioner.
  • the operation data of the device 2 is time-series data representing the operation state of the device 2.
  • the insurance premium rate is the ratio of the insurance premium to the insurance amount.
  • the insurance premium is the amount paid by the insured to the insurance company.
  • Insurance premiums are calculated by multiplying the insurance premium rate by the insurance amount.
  • the insurance amount is an amount received by an insured person such as the owner of the device 2.
  • the insurance setting support system 1 obtains the insurance premium rate set for the device 2 and proposes the insurance premium rate calculated from the insurance premium rate to the insurance company.
  • the insurance setting support system 1 includes an insurance premium rate calculation device 3 and an output device 4.
  • the insurance premium rate calculation device 3 estimates the future state of the device 2 by statistically analyzing the operation data of the device 2, and calculates the insurance premium rate for the device 2 based on the future state of the device 2.
  • the future state of the device 2 is, for example, the remaining life of the device 2 or the failure rate of the device 2.
  • the remaining life of the device 2 is the time from the current time until the device 2 becomes inoperable.
  • the failure rate of the device 2 is the probability that the device 2 will fail.
  • the remaining life of the device 2 and the failure rate of the device 2 are obtained for each failure type.
  • the future state of the device 2 is not limited to the future state indicated by the remaining life and the failure rate. For example, the degree of deterioration indicating the degree of change from the initial operating state of the device 2 although the device 2 does not fail. May be used.
  • the future state of the device 2 is generally different depending on the initial state or the operating environment of the device 2, even if the devices 2 have the same model and the same function.
  • the insurance premium rate calculation device 3 estimates the future state for each device 2 by statistically analyzing the operation data obtained from each device 2, and the individual device 3 is based on the estimated future state for each device 2. Calculate the insurance premium rate for the device 2. As a result, the insurance setting support system 1 provided with the insurance premium rate calculation device 3 can propose an appropriate insurance plan corresponding to each device 2.
  • the output device 4 outputs insurance proposal information including the insurance premium rate of the device 2 calculated by the insurance premium rate calculation device 3 and the insurance plan determined by using this insurance premium rate.
  • the output device 4 is a Web server.
  • the Web server receives the insurance proposal information of the device 2 from the insurance premium rate calculation device 3, and stores the insurance proposal information for each device 2 in the memory.
  • a terminal device equipped with a Web browser such as a smartphone, a tablet terminal, or a PC (Personal Computer) can view insurance proposal information from the Web browser.
  • the insurance premium rate calculation device 3 includes a data acquisition unit 31, a device state estimation unit 32, an insurance premium rate calculation unit 33, an output processing unit 34, and an insurance premium rate database 35.
  • the data acquisition unit 31 acquires the operation data of the device 2.
  • the data acquisition unit 31 acquires operation data from the control circuit in the device 2 or the sensor mounted on the device 2.
  • the operation data includes, for example, the operation mode of the air conditioner (cooling, heating, ventilation, etc.), the temperature of the heat exchanger, the frequency of the compressor, and the rotation speed of the compressor. Is done.
  • the device state estimation unit 32 estimates the future state of the device 2 by statistically analyzing the operation data of the device 2.
  • Methods for statistically analyzing operational data include autoregressive models or support vector regression.
  • the autoregressive model is a learning model that regresses and estimates an estimated value indicating the state of the device 2 at a certain time using operation data before this time.
  • the support vector regression estimates the future state of the device 2 by regressing using the data (support vector) necessary for estimating the state of the device 2 from the past operation data.
  • the insurance premium rate calculation unit 33 calculates the insurance premium rate for the device 2 based on the future state of the device 2. For example, the insurance premium rate calculation unit 33 calculates the insurance premium rate for the device 2 by using a function related to the insurance premium rate with the data indicating the future state of the device 2 as a variable. Further, the insurance premium rate calculation unit 33 may select the insurance premium rate from the insurance premium rate database 35 by using the data indicating the future state of the device 2 estimated by the device state estimation unit 32.
  • the output processing unit 34 performs a process of outputting the insurance premium rate for the device 2 calculated by the insurance premium rate calculation unit 33 to the output device 4. Further, the output processing unit 34 may output the insurance plan of the insurance premium calculated based on the insurance premium rate in addition to the insurance premium rate for the device 2. Further, the output processing unit 34 may output to the output device 4 an insurance plan that can be set in the device 2 based on the confidence interval of the estimation of the remaining life or the failure rate of the device 2 by the device state estimation unit 32.
  • the insurance premium rate database 35 is a database in which data indicating the future state of the device 2 and the insurance premium rate corresponding to this data are registered.
  • the data indicating the future state of the device 2 is, for example, the remaining life of the device 2 or the failure rate of the device 2.
  • the correspondence between the remaining life or failure rate and the insurance premium rate is set by statistically analyzing the remaining life or failure rate and the insurance premium rate.
  • the insurance premium rate database 35 is provided for each device 2 even if the model and function of the device 2 are the same.
  • the insurance premium rate database 35 may be a database owned by an external storage device provided separately from the insurance premium rate calculation device 3. In this case, the insurance premium rate calculation device 3 does not have the insurance premium rate database 35, and the insurance premium rate calculation unit 33 acquires information indicating the insurance premium rate from the insurance premium rate database 35 of the external storage device by wired communication or wireless communication. Will be done.
  • FIG. 2 is a flowchart showing the insurance premium rate calculation method according to the first embodiment.
  • the data acquisition unit 31 sequentially acquires the operation data of the device 2 from the control circuit in the device 2 or the sensor mounted on the device 2 (step ST1).
  • the operation data of the air conditioner includes, for example, the operation mode of the air conditioner (heating, cooling, ventilation, set temperature, etc.), the number of revolutions of the compressor, the frequency of the compressor, or the heat exchanger. Temperature is included.
  • the data acquisition unit 31 may acquire weather data or area information around the installation location of the device 2 in addition to the operation data of the device 2.
  • Meteorological data includes, for example, outside air temperature, weather or humidity.
  • the area information is, for example, information indicating whether the place where the device 2 is installed is a coastal area or a mountainous area.
  • the data acquisition unit 31 may acquire the installation condition information of the device 2.
  • the installation condition information of the device 2 is information indicating the installation state of the device 2, and is, for example, the number of indoor units, the pipe length, or the amount of refrigerant charged when the device 2 is an air conditioner.
  • the device state estimation unit 32 estimates information indicating the future state of the device 2 using the data acquired by the data acquisition unit 31 (step ST2).
  • FIG. 3 is an explanatory diagram showing an outline of a process for estimating a future state of the device 2.
  • the operation data A1 of the device 2 acquired by the data acquisition unit 31 is time-series data of various information indicating the operation state of the device 2.
  • the device state estimation unit 32 calculates the time-series data of the abnormality degree B1 past the current time T1 of the device 2 by using the operation data A1 of the device 2.
  • the degree of abnormality takes a value of 0 or more according to the degree of abnormality that has occurred in the device 2.
  • the larger the value of the degree of abnormality the higher the degree of abnormality of the device 2.
  • the device state estimation unit 32 calculates the past abnormality degree B1 of the device 2 by using, for example, a physical model formula expressing the degree of abnormality using the operation data of the device as a variable.
  • the abnormality degree B1 is a value that differs depending on the model or model of the device 2.
  • the model is the product model of the device 2.
  • the device state estimation unit 32 uses a relational expression that calculates the degree of abnormality of the device 2 for each model or model by inputting the operation data of the device 2 as a variable, and the device 2 is used for each model or model.
  • the degree of abnormality B1 may be calculated.
  • the degree of abnormality is calculated by inputting a numerical value uniquely assigned to the model or model of the device 2 into the variable of the above function.
  • the above function is derived, for example, by analyzing the time-series data of the operation data of the device 2 for each model or model.
  • the abnormality degree B1 is a value that differs depending on the type of failure that occurs in the device 2. Therefore, the device state estimation unit 32 inputs the operation data of the device 2 as a variable, and uses a relational expression for calculating the degree of abnormality caused by the failure that occurred in the device 2, for each type of failure of the device 2.
  • the degree of abnormality B1 may be calculated.
  • the air conditioner causes failures peculiar to the air conditioner, such as leakage of refrigerant, deterioration of the compressor, and corrosion of the heat exchanger.
  • the equipment state estimation unit 32 calculates the abnormality degree B1 caused by various failures by using the relational expressions corresponding to the leakage of the refrigerant, the deterioration of the compressor, and the corrosion of the heat exchanger.
  • the equipment state estimation unit 32 calculates the abnormality degree B1 (refrigerant leakage degree) caused by the refrigerant leakage generated in the equipment 2 by inputting the operation data of the equipment 2 into the variable of the relational expression relating to the refrigerant leakage.
  • the equipment state estimation unit 32 uses the relational expression regarding the deterioration of the compressor to determine the abnormality degree B1 (degree of deterioration of the compressor) caused by the deterioration of the compressor. Calculate and use the relational expression related to heat exchanger corrosion to calculate the degree of abnormality B1 (heat exchanger corrosion degree) caused by the corrosion of the heat exchanger.
  • the device state estimation unit 32 uses a relational expression with at least one of meteorological data, area information, and installation condition information as a variable in addition to the operation data of the device 2, and determines the past abnormality degree B1 of the device 2. It may be calculated.
  • the data acquisition unit 31 acquires the weather data and the area information from the Web server that manages the weather data or the area information, and acquires the installation condition information from the terminal of the inspection worker of the device 2.
  • Meteorological data such as outside air temperature, weather or humidity is information that affects the operating condition of the air conditioner.
  • the outside air temperature is a factor that influences the coefficient of performance representing the cooling capacity or the heating capacity per 1 kW of power consumption of the air conditioner. Therefore, the device state estimation unit 32 accurately assumes the deterioration state of the operation capacity of the device 2 by using the relational expression for calculating the abnormality degree B1 using the data obtained by quantifying the operation data and the meteorological data of the device 2 as variables. It is possible to calculate the abnormal degree B1.
  • the data obtained by quantifying the meteorological data may be a numerical value uniquely assigned to each weather such as sunny weather, cloudy weather, or rainy weather, in addition to a specific numerical value such as outside air temperature or humidity.
  • the device state estimation unit 32 accurately assumes the deterioration state of the operation capacity of the device 2 by using the relational expression for calculating the abnormality degree B1 using the data obtained by quantifying the operation data and the area information of the device 2 as variables. It is possible to calculate the degree of abnormality B1.
  • the data obtained by quantifying the area information may be a value uniquely assigned to each area such as a coastal area or a mountainous area.
  • the device state estimation unit 32 estimates the future abnormality degree B2 of the device 2 as shown by the arrow (2) in FIG.
  • the device state estimation unit 32 inputs a past operation data A2 of the device 2 acquired by the data acquisition unit 31 and a past abnormality degree B1, and outputs a learning model that outputs a future abnormality degree B2 of the device 2. It is used to estimate the degree of abnormality B2 (time-series data indicated by the broken line) after (future) the current time T1 of the device 2. As a result, the time series of the future abnormality degree B2 of the device 2 is estimated from the tendency of the time series of the past abnormality degree B1.
  • the equipment state estimation unit 32 estimates not only the tendency of the past abnormality degree B1 but also the tendency of the past operation data A2 of the equipment 2 in the estimation of the abnormality degree B2. Is also considered. As a result, the device state estimation unit 32 can accurately estimate the future abnormality degree B2 of the device 2.
  • the learning model is, for example, an autoregressive model or a support vector regression trained to output the future abnormality degree B2 of the device 2 by inputting the past operation data A2 of the device 2 and the past abnormality degree B1. It is a model.
  • An autoregressive or support vector regression model is a machine learning model trained using known regression algorithms such as autoregressive or support vector regression.
  • the abnormality degree B2 has a different value depending on the model or model of the device 2 as in the abnormality degree B1. Therefore, the device state estimation unit 32 inputs the past operation data A2 of the device 2 and the past abnormality degree B1 corresponding to the model or model of the device 2, and determines the abnormality degree corresponding to the model or model of the device 2. Using the learning model to be output, the future abnormality degree B2 of the device 2 corresponding to the model or model of the device 2 may be estimated.
  • the learning model is, for example, a machine learning model trained using time-series data of operation data of the device 2 for each model or model as training data.
  • the abnormality degree B2 has a different value depending on the type of failure occurring in the device 2 as in the abnormality degree B1. Therefore, the device state estimation unit 32 inputs a past operation data A2 of the device 2 and a past abnormality degree B1 corresponding to the type of failure of the device 2, and outputs a learning model corresponding to the type of failure. It may be used to estimate the future abnormality degree B2 of the device 2 corresponding to the type of failure. For example, when the device 2 is an air conditioner, the device state estimation unit 32 uses a learning model corresponding to each of the leakage of the refrigerant, the deterioration of the compressor, and the corrosion of the heat exchanger, and the abnormality corresponding to various failures. Estimate degree B2.
  • the device state estimation unit 32 causes the refrigerant leakage generated in the device 2 by inputting the past operation data A2 of the device 2 and the abnormality degree B1 due to the past refrigerant leakage into the learning model regarding the refrigerant leakage. Estimate the future abnormality degree B2 (refrigerant leakage degree). Similarly, for the deterioration of the compressor and the corrosion of the heat exchanger, the device state estimation unit 32 uses the learning model for the deterioration of the compressor to determine the abnormality degree B2 (degree of deterioration of the compressor) caused by the deterioration of the compressor. Estimate and use the learning model for heat exchanger corrosion to estimate the degree of anomaly B2 (heat exchanger corrosion) caused by heat exchanger corrosion.
  • the device state estimation unit 32 inputs at least one of meteorological data, area information, or installation condition information in addition to the past operation data A2 and the past abnormality degree B1 of the device 2, and outputs the abnormality degree B2.
  • the future abnormality degree B2 of the device 2 may be estimated by using the learning model.
  • the device state estimation unit 32 uses a learning model that inputs the past operation data A2 of the device 2, the past abnormality degree B1 and the data obtained by quantifying the meteorological data, and outputs the abnormality degree B2. It is possible to calculate the degree of abnormality B2 in which the deterioration state of the operating capacity of 2 is accurately assumed. Further, the device state estimation unit 32 inputs the past operation data A2 of the device 2, the past abnormality degree B1 and the data obtained by quantifying the area information, and outputs the abnormality degree B2 by using the learning model. It is possible to calculate the degree of abnormality B2 in which the deterioration state of the operating capacity of 2 is accurately assumed.
  • the device state estimation unit 32 determines the remaining life and failure rate of the device 2 based on the past abnormality degree B1 and the future abnormality degree B2 of the device 2. Calculated as information indicating the future status. For example, the device state estimation unit 32 estimates the future time T2 at which the future abnormality degree B2 of the device 2 becomes the threshold value Th or more as the life of the device 2, that is, the time during which the device 2 cannot operate normally. Then, the device state estimation unit 32 calculates the difference time from the current time T1 to the time T2. The time of this difference is the remaining life of the device 2.
  • the threshold value Th is the abnormality degree B2 of the device 2 at the estimated time when the device 2 cannot operate normally.
  • the threshold value Th may be a value determined based on the output upper limit value of the sensor that detects the operation data A2 of the device 2, or is a value calculated by statistically analyzing the operation data A2 of the device 2. But it may be.
  • the device state estimation unit 32 calculates the future failure rate Fr of the device 2 by dividing each abnormality degree B2 from the current time T1 by the abnormality degree B2 (threshold value Th) of the life time of the device 2.
  • the failure rate Fr varies depending on the value of the future abnormality degree B2 of the device 2. For example, when the abnormality degree B2 is 0, the failure rate Fr becomes 0%.
  • the remaining life and failure rate of the device 2 will differ depending on the type of failure that occurs in the device 2.
  • the device state estimation unit 32 may output the remaining life and the failure rate for each type of failure to the insurance premium rate calculation unit 33, respectively, or the remaining life obtained by integrating the remaining life and the failure rate for each type of failure. And the failure rate may be output to the insurance premium rate calculation unit 33.
  • the remaining life and failure rate obtained by integrating the remaining life and failure rate for each type of failure are, for example, the average remaining life obtained by averaging the remaining life for each type of failure and the failure rate for each type of failure. It may be the average failure rate obtained by averaging. Further, the remaining life obtained by integrating the remaining life for each type of failure may be the minimum value (shortest remaining life) of the remaining life for each type of failure. Further, the failure rate obtained by integrating the failure rates for each type of failure may be the maximum value of the failure rates among the failure rates for each type of failure.
  • the future abnormality degree B2 for each device 2 may be a different value depending on the initial state of the device 2 or the operating environment of the device 2, even if the devices 2 have the same model and the same function. be. In this case, the remaining life and the failure rate also have different values for each device 2. Therefore, the device state estimation unit 32 estimates the future abnormality degree B2 for each device 2 subject to insurance setting, and estimates the remaining life and the failure rate for each type of failure. Thereby, an appropriate remaining life and failure rate corresponding to each device 2 can be estimated for each type of failure.
  • the insurance premium rate calculation unit 33 calculates the insurance premium rate for the device 2 based on the information indicating the future state of the device 2 estimated by the device state estimation unit 32 (step ST3). For example, the insurance premium rate calculation unit 33 selects the insurance premium rate corresponding to the device 2 from the insurance premium rate database 35 by using the remaining life of the device 2 estimated by the device state estimation unit 32 or the failure rate of the device 2. Further, the insurance premium rate calculation unit 33 may calculate the insurance premium rate for the device 2 by using a relational expression with the remaining life of the device 2 or the failure rate of the device 2 as a variable. As the function formula, for example, there is a linear function in which the insurance premium rate changes linearly with respect to the remaining life or the failure rate.
  • FIG. 4A is a graph showing the correspondence between the remaining life of the device 2 and the insurance premium rate.
  • FIG. 4B is a graph showing the correspondence between the failure rate of the device 2 and the insurance premium rate.
  • 4A and 4B show a case where it is assumed that the remaining life of the device 2 and the failure rate of the device 2 and the insurance premium rate set in the device 2 have a linear function relationship.
  • the insurance premium rate database 35 the correspondence between the remaining life and the failure rate and the insurance premium rate is registered for each device 2. Further, the correspondence between the remaining life and the failure rate and the insurance premium rate may be registered in the insurance premium rate database 35 for each device 2 and for each type of failure.
  • the insurance premium rate calculation unit 33 selects the insurance premium rate for the device 2 by referring to the correspondence shown in FIG. 4A using the information indicating the remaining life of the device 2 estimated by the device state estimation unit 32. .. Further, the insurance premium rate calculation unit 33 selects the insurance premium rate for the device 2 by referring to the correspondence shown in FIG. 4B using the information indicating the failure rate of the device 2 estimated by the device state estimation unit 32. .. For example, if the remaining life of the device 2 is long (or the failure rate is low), a low insurance premium rate is set for the device 2 in order to recover profits over a long period of time. Further, when the remaining life of the device 2 is short (or the failure rate is high), a high insurance premium rate is set for the device 2 in order to recover the profit in a short period of time.
  • the output processing unit 34 outputs the insurance premium rate for the device 2 to the output device 4 (step ST4).
  • the output device 4 holds information indicating the insurance premium rate for the device 2 so that it can be viewed by a Web browser.
  • a Web browser For example, before proposing an insurance plan to an insurance company, it is possible to browse and examine information indicating an insurance premium rate for device 2 from a Web browser installed in a PC. It is also possible for the customer to browse the insurance premium rate for the device 2 from a Web browser mounted on a smartphone or tablet terminal.
  • the output processing unit 34 may output insurance proposal information for the device 2 to the output device 4.
  • the insurance proposal information is information including the insurance premium rate for the device 2 and a plurality of insurance plans that can be set for the device 2.
  • the remaining life or failure rate is generally different depending on the type of failure even for the same device 2. Therefore, the insurance premium rate calculation unit 33 calculates the insurance premium rate for each type of failure that may occur in the device 2.
  • the output processing unit 34 outputs insurance proposal information including a plurality of insurance plans according to the insurance premium rate for each type of failure to the output device 4. As a result, the user can set an appropriate insurance plan from a plurality of insurance plans that can be set for the device 2.
  • the output processing unit 34 generates a stepwise insurance plan based on the confidence interval of estimating the remaining life or the failure rate of the device 2, and outputs the insurance proposal information including the stepwise insurance plan to the output device 4. It may be output.
  • the device state estimation unit 32 estimates the remaining life of the device 2 in various future periods, obtains the normal distribution of the estimated values of the remaining life, and calculates the confidence interval of the remaining life estimation.
  • the output processing unit 34 outputs an insurance plan in which the insurance premium rate is set stepwise based on the confidence interval of the remaining life estimation calculated by the device state estimation unit 32.
  • the output processing unit 34 outputs an insurance plan with a high premium rate when the remaining life of the device 2 is shorter than the average value ⁇ in each confidence interval, and when the remaining life of the device 2 is longer than the average value ⁇ . Outputs an insurance plan with a low premium rate.
  • proposing an insurance plan for the device 2 based on the confidence interval of the remaining life is shown, it is also possible to propose an insurance plan for the device 2 based on the confidence interval of the failure rate of the device 2.
  • the device state estimation unit 32 may estimate the current state of the device 2 by statistically analyzing the past operation data up to the current time.
  • the insurance premium rate calculation unit 33 calculates the insurance premium rate for the device 2 based on the information indicating the current state of the device 2.
  • the output processing unit 34 outputs an insurance plan based on the insurance premium rate according to the current state of the device 2 to the output device 4. For example, if the degree of deterioration of the current device 2 is high, it is estimated that the remaining life of the device 2 is short, so an insurance plan with a high premium rate is proposed. Further, if the degree of deterioration of the current device 2 is low, an insurance plan with a low premium rate is proposed. This makes it possible to set an insurance plan according to the current state of the device 2.
  • the device state estimation unit 32 may update the learning model for estimating the future state of the device 2. For example, the device state estimation unit 32 uses the learning model of the first device of interest among the plurality of devices 2 subject to insurance setting, and the learning model of the second device different from the first device. Update based on the estimated anomaly degree B2. For example, when the first device and the second device are of the same model and have similar installation conditions, it is preferable that the first device and the second device have similar insurance plans. Therefore, the device state estimation unit 32 reduces the value of the difference between the abnormality degree B2 estimated by using the learning model in the first device and the abnormality degree B2 estimated by using the learning model in the second device. , Update the parameters of the learning model in the first device.
  • the device state estimation unit 32 may update the learning model in the device 2 before and after the device 2 is repaired. It is considered that the device 2 repaired in the periodic inspection or the like operates in a state close to the initial stage after the deterioration over time before the repair is eliminated. In this case, it is highly possible that the learning model in the device 2 before the repair cannot accurately estimate the future state of the device 2 after the repair.
  • the device state estimation unit 32 resets the learning data for the learning model in the device 2 before the repair, and relearns the learning model using the operation data obtained from the device 2 after the repair, for example.
  • the device 2 after repair has a longer remaining life and a lower failure rate than the device 2 before repair. Therefore, by updating the learning model in the device 2 after repair, the remaining life or the failure rate can be accurately estimated.
  • the device 2 after repair can be set with a lower insurance premium rate than the device 2 before repair.
  • FIG. 5A is a block diagram showing a hardware configuration that realizes the function of the insurance premium rate calculation device 3.
  • FIG. 5B is a block diagram showing a hardware configuration for executing software that realizes the function of the insurance premium rate calculation device 3.
  • the processing circuit 104 or the processor 105 acquires operation data from the device 2 side via the input interface 100. Further, the processing circuit 104 or the processor 105 outputs the insurance proposal information to the output device 4 via the output interface 101.
  • the insurance premium rate calculation unit 33 shown in FIG. 1 refers to the storage contents of the storage device 103 via the storage device interface 102.
  • the storage device 103 stores the insurance premium rate database 35.
  • the components shown in FIGS. 5A and 5B are connected to each other by a signal line.
  • the processing circuit 104 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, or an ASIC (Application Specific Integrated Circuitd). Circuit), FPGA (Field-Programmable Gate Array), or a combination of these is applicable.
  • the functions of the data acquisition unit 31, the device state estimation unit 32, the insurance premium rate calculation unit 33, and the output processing unit 34 provided in the insurance premium rate calculation device 3 may be realized by separate processing circuits, or these functions may be collectively realized. It may be realized by one processing circuit.
  • the processing circuit is the processor 105 shown in FIG. 5B
  • the functions of the data acquisition unit 31, the device state estimation unit 32, the insurance rate calculation unit 33, and the output processing unit 34 included in the insurance rate calculation device 3 are software, firmware, or It is realized by a combination of software and firmware.
  • the software or firmware is described as a program and stored in the memory 106.
  • the processor 105 By reading and executing the program stored in the memory 106, the processor 105 functions as a data acquisition unit 31, a device state estimation unit 32, an insurance rate calculation unit 33, and an output processing unit 34 included in the insurance rate calculation device 3.
  • the insurance premium rate calculation device 3 includes a memory 106 for storing a program in which the processes of steps ST1 to ST4 in the flowchart shown in FIG. 2 are executed as a result when executed by the processor 105.
  • These programs cause a computer to execute the procedures or methods of the data acquisition unit 31, the device state estimation unit 32, the insurance premium rate calculation unit 33, and the output processing unit 34.
  • the memory 106 may be a computer-readable storage medium in which a program for making the computer function as a data acquisition unit 31, a device state estimation unit 32, an insurance premium rate calculation unit 33, and an output processing unit 34 is stored.
  • the memory 106 may be, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically-volatile) or an EEPROM (Electrically-EPROM).
  • RAM Random Access Memory
  • ROM Read Only Memory
  • flash memory an EPROM (Erasable Programmable Read Only Memory)
  • EEPROM Electrically-volatile
  • EEPROM Electrically-EPROM
  • the functions of the data acquisition unit 31, device state estimation unit 32, insurance premium rate calculation unit 33, and output processing unit 34 included in the insurance premium rate calculation device 3 are realized by dedicated hardware, and the remaining part is software or It may be realized by firmware.
  • the function of the data acquisition unit 31 is realized by the processing circuit 104, which is dedicated hardware, and the processor 105 of the device state estimation unit 32, the insurance premium rate calculation unit 33, and the output processing unit 34 is stored in the memory 106.
  • the function is realized by reading and executing the program.
  • the processing circuit can realize the above-mentioned functions by hardware, software, firmware or a combination thereof.
  • the insurance premium rate calculation device 3 is the data acquisition unit 31 that acquires the operation data of the device 2, and the future of the device 2 by statistically analyzing the operation data of the device 2.
  • the device state estimation unit 32 for estimating the state and the insurance premium rate calculation unit 33 for calculating the insurance premium rate for the device 2 based on the information indicating the future state of the device 2 are provided. Since the future state of the device 2 is estimated using the operation data of the device 2, the insurance premium rate calculation device 3 can propose the insurance premium rate corresponding to the device 2.
  • the insurance premium rate calculation device can be used, for example, for setting insurance for an air conditioner.
  • 1 insurance setting support system 2 equipment, 3 insurance premium rate calculation device, 4 output device, 31 data acquisition unit, 32 equipment status estimation unit, 33 insurance premium rate calculation unit, 34 output processing unit, 35 insurance premium rate database, 100 input interface, 101 output interface, 102 storage device interface, 103 storage device, 104 processing circuit, 105 processor, 106 memory.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

Le dispositif de calcul de taux de prime d'assurance (3) d'après la présente invention comprend : une unité d'acquisition de données (31) qui acquiert des données de fonctionnement indiquant un état de fonctionnement d'un appareil (2) ; une unité d'estimation d'état d'appareil (32) qui estime un futur état de l'appareil (2) au moyen d'une analyse statistique des données de fonctionnement de l'appareil (2) ; et une unité de calcul de taux de prime d'assurance (33) qui calcule un taux de prime d'assurance pour l'appareil (2) sur la base des informations indiquant le futur état de l'appareil (2).
PCT/JP2020/041003 2020-11-02 2020-11-02 Dispositif et procédé de calcul de taux de prime d'assurance WO2022091399A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/041003 WO2022091399A1 (fr) 2020-11-02 2020-11-02 Dispositif et procédé de calcul de taux de prime d'assurance

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/041003 WO2022091399A1 (fr) 2020-11-02 2020-11-02 Dispositif et procédé de calcul de taux de prime d'assurance

Publications (1)

Publication Number Publication Date
WO2022091399A1 true WO2022091399A1 (fr) 2022-05-05

Family

ID=81382156

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/041003 WO2022091399A1 (fr) 2020-11-02 2020-11-02 Dispositif et procédé de calcul de taux de prime d'assurance

Country Status (1)

Country Link
WO (1) WO2022091399A1 (fr)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003248757A (ja) * 2002-12-19 2003-09-05 Aiu Insurance Co 契約証書発行システム
JP2006155008A (ja) * 2004-11-26 2006-06-15 Shigemi Taro 土木建築機械の貸借システム及び土木建築機械の保険システム
WO2018181020A1 (fr) * 2017-03-29 2018-10-04 三菱重工業株式会社 Système de détection d'indicateur et procédé de détection d'indicateur
WO2019116450A1 (fr) * 2017-12-12 2019-06-20 本田技研工業株式会社 Système de définition de prime d'assurance
JP2019160128A (ja) * 2018-03-16 2019-09-19 株式会社日立製作所 故障確率評価システム及び方法
JP2020166324A (ja) * 2019-03-28 2020-10-08 栗田工業株式会社 保険契約料金算出装置及び保険契約料金算出方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003248757A (ja) * 2002-12-19 2003-09-05 Aiu Insurance Co 契約証書発行システム
JP2006155008A (ja) * 2004-11-26 2006-06-15 Shigemi Taro 土木建築機械の貸借システム及び土木建築機械の保険システム
WO2018181020A1 (fr) * 2017-03-29 2018-10-04 三菱重工業株式会社 Système de détection d'indicateur et procédé de détection d'indicateur
WO2019116450A1 (fr) * 2017-12-12 2019-06-20 本田技研工業株式会社 Système de définition de prime d'assurance
JP2019160128A (ja) * 2018-03-16 2019-09-19 株式会社日立製作所 故障確率評価システム及び方法
JP2020166324A (ja) * 2019-03-28 2020-10-08 栗田工業株式会社 保険契約料金算出装置及び保険契約料金算出方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
IMAMURA, MAKOTO: "A Survey on Machine Learning for Prognostics and Health Management ", INSTITUTE OF ELECTRICAL ENGINEERS OF JAPAN (IEEJ) STUDY GROUP MATERIALS, vol. 2016, no. 49-64, 28 October 2016 (2016-10-28), pages 51 - 56, XP009537456 *

Similar Documents

Publication Publication Date Title
AU2022201272B2 (en) Maintenance and diagnostics for refrigeration systems
US20200356087A1 (en) Model predictive maintenance system with event or condition based performance
US20200301408A1 (en) Model predictive maintenance system with degradation impact model
US20200408447A1 (en) Energy Management for Refrigeration Systems
US10592821B2 (en) Self-learning fault detection for HVAC systems
TWI709092B (zh) 故障機率評估系統及方法
US11371734B2 (en) Method and computer system for monitoring an HVAC system with data from a plurality of HVAC controllers
US7966152B2 (en) System, method and algorithm for data-driven equipment performance monitoring
US9477224B2 (en) Physics-based lifespan modeling
US11774122B2 (en) Building control system with adaptive online system identification
WO2021026370A1 (fr) Système de maintenance prédictive de modèle avec modèle d'impact de dégradation
JP6862130B2 (ja) 異常検知装置、異常検知方法、およびプログラム
WO2022091399A1 (fr) Dispositif et procédé de calcul de taux de prime d'assurance
US20210190354A1 (en) Building system with early fault detection
US9507344B2 (en) Index generation and embedded fusion for controller performance monitoring
JP6986358B2 (ja) 情報処理装置、情報処理方法およびプログラム
US20230349608A1 (en) Anomaly detection for refrigeration systems
JP7134915B2 (ja) プログラム、クレーム発生予測装置及びクレーム発生予測方法
AU2022204337B2 (en) Run-time reliability reporting for electrical hardware systems
US20210398003A1 (en) Method and system for forecasting a failure of a ventilator group, and corresponding ventilator group
JP2023120803A (ja) 機器状態予測方法及び機器状態予測システム
CN113901294A (zh) 一种信息处理方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20959914

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20959914

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP