JP5355341B2 - Apparatus and method for predicting occurrence of insurance accident based on disaster - Google Patents

Description

  The present invention relates to a technique for predicting the occurrence of an insurance accident due to a disaster.

  When a policyholder or an insured person encounters an insurance accident, the insurance company often knows that the insurance accident has occurred for the first time after receiving an accident report from the policyholder or the like. Therefore, in general, operations such as investigation and payment of insurance money are generally started by a claim for insurance money from an insured or the like.

  By the way, there is a technology for predicting damage caused by a disaster. For example, Patent Document 1 describes a technique for predicting damage to a building due to an earthquake, and Patent Document 2 describes a technique for predicting damage due to water damage.

JP 2009-115650 A JP 2006-4212 A

  However, it is rare for ordinary consumers to read the policy clauses of their insurance contracts, and there are almost no policyholders who fully understand the details of insurance coverage they have. Therefore, even if an insurance accident that is subject to compensation occurs, some contractors are unaware of it and do not make insurance claims. The so-called insurance non-payment problem is said to be partly due to the fact that the policyholder did not make a claim for payment.

  In order to eliminate this non-payment of insurance claims, measures such as close communication between insurance companies and policyholders can be considered, but ultimately, a system for grasping active insurance accidents by insurance companies is established. Is essential. For example, among insurance accidents, in particular, in the case of an insurance accident caused by a widespread disaster including natural disasters and man-made disasters, active understanding based on damage prediction by an insurance company is possible.

  Therefore, an object of the present invention is to provide a technique for predicting the occurrence of an insurance accident due to a disaster.

  Another object of the present invention is to provide a technique for an insurer to actively grasp the occurrence of an insurance accident even when there is no request from an insurance policyholder or the like.

  According to one embodiment of the present invention, a disaster-based insurance accident occurrence prediction device accepts input of event data relating to a disaster-causing event and predicts the degree of disaster damage by district, and an insured person Insurance contract storage means for storing insurance contract data relating to insurance contracts by insurance company, including the residence of the insurer, and based on the injuries of the insured person and the extent of damage by district predicted by the damage prediction means A degree of damage predicted by the insured person, a means for connecting to a plurality of insurance company terminal devices, and a degree of damage for each insured predicted by the degree of damage predicted means is predetermined. Extraction means for extracting an insurance contract relating to the insured person as described above. The connection means notifies the insurance contract extracted by the extraction means to the insurance company terminal device of the insurance company related to the insurance contract.

  In a preferred embodiment, the insurance contract data includes an insured's residence floor, the event data is precipitation data for each time zone, and the damage prediction means indicates topography and geology. Having land attribute data, predicting the water level for each area based on the land attribute data and the precipitation data for each time zone, and the disaster degree predicting means includes the water level for each area and the water level for each area. The degree of damage per insured may be predicted based on the insured's residence and floor.

  In a preferred embodiment, the insurance contract data includes an insured's residential building attribute, the event data includes data indicating an earthquake epicenter and an earthquake magnitude, and the damage prediction means includes a geological feature. The land attribute data, and based on the land attribute data, the magnitude of the earthquake, and the distance from the epicenter of the earthquake to each district, the earthquake is predicted for each region, and the degree of damage prediction means May predict the degree of damage for each insured based on the shaking for each area predicted by the damage prediction means and the residential building attribute of the insured.

It is a lineblock diagram of an insurance accident prediction system concerning one embodiment of the present invention. An example of the data structure of an insurance contract data storage part is shown. An example of an earthquake damage rate curve is shown. It is a flowchart which shows the process sequence of insurance accident occurrence prediction by flood. It is a flowchart which shows the process sequence of insurance accident occurrence prediction by an earthquake disaster.

  Hereinafter, an insurance accident prediction system according to an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a configuration diagram of an insurance accident prediction system according to the present embodiment. As shown in the figure, this system includes an insurance accident prediction device 1 that predicts the occurrence of an insurance accident due to a widespread disaster, and a plurality of insurance companies connected to the insurance accident prediction device 1 via a network. Terminal devices 3, 3, ... are provided.

  The insurance accident prediction device 1 and the insurance company-specific terminal devices 3, 3,... Are each constituted by, for example, a general-purpose computer system, and the insurance accident prediction device 1 and the insurance company-specific terminal devices 3, 3 described below. ,... Are realized by executing a computer program, for example. This computer program can be stored in a computer-readable recording medium.

  The insurance accident prediction apparatus 1 predicts the occurrence of an insurance accident due to a disaster that can be covered by an insurance contract. In the present invention, the target for predicting the occurrence of an insurance accident is a widespread disaster. Therefore, in the following embodiment, an earthquake disaster (collapse) and flood damage (rainfall) will be described as an example. In other words, in this embodiment, the insurance accident prediction apparatus 1 includes a flood compensation prediction system 10 that predicts the occurrence of an insurance accident that can be compensated by flood damage, and an earthquake disaster that predicts the occurrence of an insurance accident that can be compensated by an earthquake disaster. The compensation prediction system 20 includes an insurance contract data storage unit 30 that stores a plurality of insurance contract data, and a network interface unit 40.

  Here, the insurance accident prediction apparatus 1 according to the present invention can be applied to a surface disaster other than a disaster other than an earthquake disaster (collapse) and flood damage (rainfall). For example, it may be a tsunami disaster, flood damage caused by a levee break, liquefaction caused by an earthquake or fire spread, fire, or a disaster caused by an explosion such as a factory or a bomb. That is, the insurance accident prediction apparatus 1 according to the present invention can be applied to both natural disasters and man-made disasters as long as the damage is widespread.

  The insurance contract data storage unit 30 stores data related to insurance contracts exchanged between a plurality of insurance companies and their respective policyholders. For example, the insurance contract data storage unit 30 stores insurance contract data relating to insurance contracts by insurance company, including the residence of the insured.

  FIG. 2 shows an example of the data structure of the insurance contract data storage unit 30. As shown in the figure, the insurance contract data storage unit 30 includes, as data items, an insurance company ID 301, a contract number 302, a contractor name 303, an insured person name 304, an insurance item 305, and a payment trigger 307. And a residential area 309, a residential floor 311, a construction year 313, a structure 315, a floor 317, an insurance amount 319, and an insurance premium 321.

  The insurance company ID 301 is identification information of an insurance company that is one party of the insurance contract. The contract number 302 is a number that uniquely identifies each insurance contract. The contractor name 303 is the name of the contractor who is the other party of the insurance contract. The insured person name 304 is the name of the insured person who is compensated by the insurance contract. The insured may be the same as the insurer. The insurance item 305 indicates items such as non-life insurance and earthquake insurance. The payment trigger 307 is a standard for determining whether to pay insurance money. Therefore, when the payment trigger 307 is exceeded, an insurance accident has occurred. The payment trigger 307 is, for example, the standard water level for non-life insurance (water damage), and the degree of building damage (total loss, half-loss, etc.) for earthquake insurance. The residence 309 is an address of a residence (building) of an insured person to be compensated. The residence floor 311 is the number of stories when the residence of the insured subject to compensation is an apartment. The construction year 313 is the year in which the insured's building to be compensated was built. The structure 315 is the structure of the insured's building to be compensated. The number of floors 317 is the number of floors of the insured's building to be compensated. The building year 313, the structure 315, and the number of floors 317 are attributes of the building to be compensated in the case of earthquake insurance.

  The flood compensation prediction system 10 includes a water level prediction unit 11, a flood damage degree prediction unit 13, and a compensation target contract extraction unit 15. The earthquake disaster prediction system 20 includes an earthquake analysis unit 21, an earthquake damage degree prediction unit 23, and a compensation target contract extraction unit 25. The insurance accident prediction apparatus 1 further includes compensation target contract storage units 17 and 27 that store the compensation target contracts extracted by the flood compensation prediction system 10 and the earthquake disaster prediction system 20, respectively.

  The water level prediction unit 11 and the earthquake analysis unit 21 are damage prediction means that accepts input of event data related to an event that causes a disaster and predicts the degree of damage due to the disaster for each district. The flood damage degree prediction unit 13 and the earthquake damage degree prediction unit 23 predict the damage degree for each insured person based on the injured person's residence and the degree of damage for each area predicted by the damage prediction means. It is a degree prediction means. The compensation target contract extraction units 15 and 25 are extraction means for extracting an insurance contract related to an insured person whose degree of damage per insured predicted by the degree of damage prediction means is equal to or greater than a predetermined value.

  The water level prediction unit 11 receives precipitation data in units of a predetermined time (for example, 10 minutes) at a number of observation points, based on the land attribute data 110 indicating the topography, geology, etc. stored in advance in the interior, Predict water levels by district. For example, the water level prediction unit 11 may predict a time change of the water level for each mesh with a 50 m square mesh as one area. The position of each mesh is specified by latitude and longitude. In the present embodiment, the water level prediction unit 11 outputs the water level for each mesh at different times. That is, the water level prediction unit 11 predicts the surface spread of the water level.

  Since the water level prediction unit 11 can use an existing water level prediction system, the detailed description of the processing is omitted.

  The flood damage degree prediction unit 13 predicts the damage degree for each insured person based on the water level for each area and the residence and floor of the insured person. For example, for each record of the insurance contract data stored in the insurance contract data storage unit 30, a mesh including the residence 309 is specified. If the residence 309 is written as an address, it is converted into latitude and longitude, and the mesh of the residence 309 is specified. Then, the flood damage degree prediction unit 13 estimates the individual inundation degree (water level) of each insured person based on the mesh water level corresponding to the residence 309 and the residence floor 311.

  The compensation target contract extraction unit 15 extracts the compensation target contract based on the inundation level for each insured person. For example, the inundation degree (water level) for each victim estimated by the flood damage degree prediction unit 13 is compared with the payment trigger 307 to extract an insurance contract that may require payment of the insurance contract.

  The storage area of the compensation target contract storage unit 17 may be divided into a plurality of areas in advance. At this time, each divided area is associated with the insurance company ID 301. The compensation target contract extraction unit 15 stores the insurance contract data extracted by the above-described process in the corresponding divided areas of the compensation target contract storage unit 17 based on the insurance company ID 301.

  When the earthquake data indicating the epicenter of the earthquake and the magnitude of the earthquake is input, the earthquake analysis unit 21 determines the distance from the epicenter and the earthquake based on the land attribute data 210 indicating the geology and the like previously stored therein. Depending on the scale of the earthquake, predict the degree of earthquake shaking in each district. For example, the earthquake analysis unit 21 may predict the shaking for each mesh with a 50 m square mesh as one area. The position of each mesh is specified by latitude and longitude. The data indicating the geology may be, for example, terrain classification in the land condition map indicating the strength of the surface ground. In this embodiment, the seismic analysis unit 21 outputs the seismic intensity for each mesh, the maximum speed of shaking, and the maximum acceleration. That is, the earthquake analysis unit 21 predicts a surface spread of the degree of shaking of the earthquake.

  In addition, since the earthquake analysis part 21 can utilize the existing earthquake analysis system, the description of the detailed process is omitted.

  The earthquake damage degree prediction unit 23 predicts the damage degree for each insured person based on the degree of shaking for each area, the residence of the insured person, and the building attributes. For example, for each record of the insurance contract data stored in the insurance contract data storage unit 30, a mesh including the residence 309 is specified. If the residence 309 is written as an address, it is converted into latitude and longitude, and the mesh of the residence 309 is specified. Then, the earthquake damage degree prediction unit 23 calculates the damage degree of each building based on the degree of shaking of the mesh corresponding to the residence 309 and the building attributes (building year 313, structure 315, and floor number 317). .

  For example, the earthquake damage degree prediction unit 23 may hold in advance data relating to a damage rate curve that indicates the degree of damage that occurs due to shaking for each structure of a building. An example of the damage rate curve is shown in FIG. The horizontal axis indicates the strength of shaking (for example, maximum acceleration, maximum speed or seismic intensity), and the vertical axis indicates the degree of damage to the building. For example, the damage rate curve is drawn for each building attribute as shown in FIG. The earthquake damage degree prediction unit 23 refers to the damage rate curve of the target building attribute and specifies the damage level (damage rate) of the building.

  For example, the compensation contract extraction unit 25 may need to pay an insurance contract based on the damage rate of the building for each insured specified by the earthquake damage prediction unit 23 and the payment trigger 307. Extract insurance contracts. Based on the insurance company ID 301, the insurance contract data is stored in 27 corresponding to each. The compensation target contract storage unit 27 is also provided with a divided area for each insurance company ID 301 in the same manner as the compensation target contract storage unit 17. The compensation target contract extraction unit 25 stores the extracted insurance contract data in the corresponding divided areas based on the insurance company ID 301.

  The network interface unit 40 is a connection means for connecting to a plurality of insurance company-specific terminal devices 3, 3,. The network interface unit 40 notifies the insurance contracts extracted by the compensation contract extraction units 15 and 25 to the insurance company terminal devices 3, 3,. As described above, the compensation contract storage units 17 and 27 store the insurance contracts extracted for each insurance company as described above. Therefore, the network interface unit 40 transmits the insurance contract data stored in the divided areas of the compensation target contract storage units 17 and 27 to the corresponding insurance company-specific terminal devices 3. For example, when the insurance accident prediction apparatus 1 is equipped with a Web server, and each insurance company-specific terminal device 3, 3,... Is equipped with a Web browser (none of which is shown), each insurance company-specific terminal device 3 is provided. , 3,... May refer to the insurance contract data of the company using a Web browser. Thereby, a plurality of insurance companies can jointly use the insurance accident prediction apparatus 1.

  Next, the processing procedure for predicting the occurrence of an insurance accident due to water damage will be described according to the flowchart shown in FIG.

  First, the water level prediction unit 11 receives input of precipitation data from an external system (S11). And the water level estimation part 11 estimates the water level according to mesh based on this received precipitation data (S13).

  Next, the flood damage degree prediction unit 13 refers to the insurance item 305 in the insurance contract data storage unit 30 and extracts a record of insurance contract data for which flood damage is to be compensated (S15). And the flood damage degree prediction part 13 specifies the water level of the insured person's residence based on the water level according to the mesh estimated at step S13 for the insurance contract of the extracted record (S17). Next, the compensation target contract extraction unit 15 determines whether the water level exceeds the payment trigger 307 based on the water level of the insured person's residence specified here and the inhabitant's residence floor 311. Determine (S19). When the water level exceeds the payment trigger 307 (S19: Yes), the target insurance contract data is stored in the compensation target contract storage unit 17 (S21). On the other hand, when the water level does not exceed the payment trigger 307 (S19: No), step S21 is skipped.

  Here, the process returns to step S15, and the flood damage degree prediction unit 13 and the compensation target contract extraction unit 15 repeat the processing from step S15 on all the insurance contract data records stored in the insurance contract data storage unit 30. Execute (S23).

  The network interface unit 40 transmits the insurance contract data stored in the compensation contract storage unit 17 for each insurance company to the corresponding insurance company terminal 3 (S25).

  According to this embodiment, it is possible to predict a flood caused by rainfall based on precipitation data and extract an insurance contract that may cause an insurance accident. As a result, the insurance company can actively grasp the occurrence of an insurance accident. Furthermore, by jointly using the insurance accident prediction apparatus 1, each insurance company can predict an insurance accident with a common specification.

  Next, the insurance accident occurrence prediction processing procedure due to the earthquake disaster will be described with reference to the flowchart shown in FIG.

  First, the earthquake analysis unit 21 receives input of earthquake data from an external system (S31). Then, the earthquake analysis unit 21 predicts the degree of earthquake vibration for each mesh based on the received earthquake data (S33).

  Next, the earthquake damage degree prediction unit 23 refers to the insurance item 305 in the insurance contract data storage unit 30 and extracts a record of insurance contract data for which the earthquake disaster is to be compensated (S35). Next, the earthquake damage degree prediction unit 23 specifies the degree of shaking of the insured's residence based on the degree of shaking for each mesh predicted in step S33 for the insurance contract of the extracted record. (S36). Then, the earthquake damage degree prediction unit 23 is based on the degree of shaking for each mesh specified here, the target building attributes (building year 313, structure 315, and floor 317), and the damage rate curve (FIG. 3). The degree of damage of the target building is predicted (S37). Next, the predicted victim extraction unit 25 determines whether or not the predicted damage level of the insured's building exceeds the payment trigger 307 (S39). When the damage level exceeds the payment trigger 307 (S39: Yes), the target insurance contract data is stored in the compensation target contract storage unit 27 (S41). On the other hand, when the damage level does not exceed the payment trigger 307 (S39: No), step S41 is skipped.

  Here, it returns to step S35 and the process after step S35 is repeated with respect to all the insurance contract data records in which the earthquake damage degree prediction part 23 and the prediction victim extraction part 25 are stored in the insurance contract data storage part 30. Execute (S43).

  The network interface unit 40 transmits the insurance contract data stored in the compensation contract storage unit 27 for each insurance company to the corresponding insurance company terminal 3 (S45).

  According to this embodiment, it is possible to predict an earthquake disaster based on earthquake data and extract an insurance contract that may cause an insurance accident.

  Then, by jointly using this system against disasters that cause surface damage such as floods and earthquake disasters, each insurance company can extract the possibility of an insurance accident from its insurance contract. it can. As a result, each insurance company can pay the insurance money promptly.

  The above-described embodiments of the present invention are examples for explaining the present invention, and are not intended to limit the scope of the present invention only to those embodiments. Those skilled in the art can implement the present invention in various other modes without departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 1 Insurance accident prediction apparatus 3,3, ... Terminal device according to insurance company 10 Flood compensation prediction system 11 Water level prediction part 13 Flood damage degree prediction part 15, 25 Compensation target contract extraction part 17, 27 Compensation target contract storage part 20 Earthquake Disaster Compensation Prediction System 21 Earthquake Analysis Unit 23 Earthquake Damage Prediction Unit 30 Insurance Contract Data Storage Unit 40 Network Interface Unit

Claims (5)

A damage prediction means that accepts input of event data related to disaster-causing events and predicts the extent of disaster damage by district;
Insurance contract storage means for storing insurance contract data relating to insurance contracts by insurance company, including the insured's residence,
Based on the injured person's residence and the degree of damage for each area predicted by the damage predicting means, the damage degree predicting means for predicting the damage degree for each insured person,
Connection means for connecting with a plurality of insurance company terminal devices;
Extracting means for extracting an insurance contract related to an insured person whose degree of damage per insured predicted by the damage degree predicting means is a predetermined value or more;
The said connection means is the occurrence prediction apparatus of an accident based on a disaster which notifies the insurance contract extracted by the said extraction means to the terminal device classified by insurance company of the insurance company which concerns on the said insurance contract. The insurance contract data includes the inhabitant's residence floor,
The event data is precipitation data for each time zone,
The damage prediction means has land attribute data indicating topography and geology, predicts the water level by district based on the land attribute data and the precipitation data by time zone by district,
The prediction of occurrence of an insurance accident based on a disaster according to claim 1, wherein the degree of damage prediction means predicts the degree of damage for each insured based on the water level for each area and the residence and floor of the insured. apparatus. The insurance contract data includes the insured's residential building attributes,
The event data includes data indicating the epicenter of the earthquake and the magnitude of the earthquake,
The damage prediction means has land attribute data including geology, and makes predictions regarding shaking by region based on the land attribute data, the magnitude of the earthquake, and the distance from the epicenter of the earthquake to each district. Done
The disaster degree predicting means is based on a disaster according to claim 1, wherein the damage degree predicting means predicts a damage degree classified by an insured person based on a swing for each area predicted by the damage predicting means and a residential building attribute of the insured person. Insurance accident occurrence prediction device. A computer having an insurance contract storage means for storing insurance contract data relating to an insurance contract by insurance company, including the insured's residence,
Accepting event data on disaster-causing events and predicting the extent of disaster damage by district;
Predicting the degree of damage by insured based on the insured's residence and the predicted extent of damage by district;
Extracting an insurance contract related to an insured person whose predicted degree of damage by insured is not less than a predetermined value;
A connection means for connecting to a plurality of insurance company terminal devices notifies the extracted insurance contract to the insurance company terminal device of the insurance company related to the insurance contract. Occurrence prediction method. A computer program for predicting the occurrence of insurance accidents based on disasters,
A computer having an insurance contract storage means for storing insurance contract data relating to insurance contracts by insurance company, including the insured's residence,
Accepting event data on disaster-causing events and predicting the extent of disaster damage by district;
Predicting the degree of damage by insured based on the insured's residence and the predicted extent of damage by district;
Extracting an insurance contract related to an insured person whose predicted degree of damage by insured is not less than a predetermined value;
A computer program for causing a connection means for connecting to a plurality of insurance company terminal devices to notify the extracted insurance contract to the insurance company terminal device of the insurance company related to the insurance contract. .

Images