JP2023084562A - Information processing device, method, program and system - Google Patents

Abstract

To estimate an amount of damage caused by a disaster at a low cost in a short time.SOLUTION: A program causes a computer to function as means of: acquiring a result of a three-dimensional survey of a disaster-stricken building damaged by the disaster, performed after an occurrence of the disaster; and estimating an amount of damage of the disaster-stricken building by applying an estimation model to input data, the estimation model constructed by using machine learning of a relation between external appearances after disasters of a plurality of buildings damaged by the disasters in the past and actual amounts of damages, based on the result of the three-dimensional survey.SELECTED DRAWING: Figure 6

Description

The present disclosure relates to an information processing device, method, program, and system.

For example, disasters such as earthquakes, fires, and landslides cause various damages to human living environments. After a disaster occurs, it is necessary to restore damaged buildings, structures, or land. It is necessary to improve the efficiency of recovery work so that victims can return to their pre-disaster lives as soon as possible.

Patent Literature 1 discloses a technique of correcting a terrain model using the result of three-dimensional measurement and calculating the total amount of damage using the flood analysis result based on the terrain model.

In Patent Document 2, three-dimensional measurement images are taken as drawing data of the original facility, and the drawing data and quantity/unit price data are linked to automatically calculate the amount of work and construction costs, and three-dimensional measurement images before and after the disaster. is disclosed to calculate the amount of recovery work by comparing the .

JP-A-2005-128838 JP-A-2006-276306

The technique of Patent Literature 1 does not assume calculation of the total damage caused by disasters other than flooding. In addition, in this technology, it is necessary to perform three-dimensional measurement before flooding occurs.

In the technique of Patent Document 2, it is necessary to take in a three-dimensional measurement image as drawing data of the prototype facility, and link the drawing data with the quantity/unit price data. In addition, this technology requires three-dimensional measurement images before and after the disaster in order to calculate the amount of recovery work.

An object of the present disclosure is to estimate the amount of damage suffered by a disaster at low cost and in a short period of time.

A program according to one aspect of the present disclosure provides a computer with means for acquiring the results of three-dimensional surveying of buildings damaged by a disaster after the occurrence of the disaster; By applying an estimation model built using machine learning on the relationship between the appearance of the building after the disaster and the actual amount of damage, to the input data based on the results of the three-dimensional survey, It functions as a means of estimating the amount of damage to a building.

1 is a block diagram showing the configuration of an information processing system according to an embodiment; FIG. It is a block diagram showing the configuration of the client device of the present embodiment. It is a block diagram which shows the structure of the server of this embodiment. 1 is an explanatory diagram of one aspect of the present embodiment; FIG. It is a figure which shows the data structure of the contract database of this embodiment. 4 is a flowchart of information processing according to the embodiment; It is a figure which shows the example of a screen displayed in the information processing of this embodiment. It is a figure which shows the example of a screen displayed in the information processing of this embodiment. It is a figure which shows the example of a screen displayed in the information processing of this embodiment. 10 is a flowchart of information processing according to Modification 1;

An embodiment of the present invention will be described in detail below with reference to the drawings. In the drawings for describing the embodiments, in principle, the same constituent elements are denoted by the same reference numerals, and repeated description thereof will be omitted.

(1) Configuration of information processing system The configuration of the information processing system will be described. FIG. 1 is a block diagram showing the configuration of the information processing system of this embodiment.

As shown in FIG. 1 , the information processing system 1 includes a client device 10 and a server 30 .
The client device 10 and server 30 are connected via a network (for example, the Internet or an intranet) NW.

The client device 10 is an example of an information processing device that transmits requests to the server 30 . The client device 10 is, for example, a smart phone, a tablet terminal, or a personal computer.

The server 30 is an example of an information processing device that provides the client device 10 with a response in response to a request transmitted from the client device 10 . Server 30 is, for example, a web server.

(1-1) Configuration of Client Apparatus The configuration of the client apparatus will be described. FIG. 2 is a block diagram showing the configuration of the client device of this embodiment.

As shown in FIG. 2, the client device 10 includes a storage device 11, a processor 12, an input/output interface 13, and a communication interface . The client device 10 is connected to the display 21 .

Storage device 11 is configured to store programs and data. The storage device 11 is, for example, a combination of ROM (Read Only Memory), RAM (Random Access Memory), and storage (eg, flash memory or hard disk).

Programs include, for example, the following programs.
・OS (Operating System) program ・Application (for example, web browser) program that executes information processing

The data includes, for example, the following data.
・Databases referenced in information processing ・Data obtained by executing information processing (that is, execution results of information processing)

The processor 12 is a computer that implements the functions of the client device 10 by activating programs stored in the storage device 11 . Processor 12 is, for example, at least one of the following:
・CPU (Central Processing Unit)
・GPU (Graphic Processing Unit)
・ASIC (Application Specific Integrated Circuit)
・FPGA (Field Programmable Array)

The input/output interface 13 is configured to acquire a user's instruction from an input device connected to the client device 10 and output information to an output device connected to the client device 10 .
Input devices are, for example, keyboards, pointing devices, touch panels, or combinations thereof.
Output devices are, for example, the display 21, speakers, or a combination thereof.

Communication interface 14 is configured to control communications between client device 10 and server 30 .

The display 21 is configured to display images (still images or moving images). The display 21 is, for example, a liquid crystal display or an organic EL display.

(1-2) Server Configuration The server configuration will be described. FIG. 3 is a block diagram showing the configuration of the server of this embodiment.

As shown in FIG. 3, the server 30 includes a storage device 31, a processor 32, an input/output interface 33, and a communication interface .

Storage device 31 is configured to store programs and data. Storage device 31 is, for example, a combination of ROM, RAM, and storage (eg, flash memory or hard disk).

Programs include, for example, the following programs.
・OS program ・Application program that executes information processing

The data includes, for example, the following data.
・Databases referenced in information processing ・Execution results of information processing

The processor 32 is a computer that implements the functions of the server 30 by activating programs stored in the storage device 31 . Processor 32 is, for example, at least one of the following:
・CPU
・GPU
・ASICs
・FPGA

The input/output interface 33 is configured to acquire user instructions from input devices connected to the server 30 and to output information to output devices connected to the server 30 .
Input devices are, for example, keyboards, pointing devices, touch panels, or combinations thereof.
An output device is, for example, a display.

Communication interface 34 is configured to control communications between server 30 and client device 10 .

(2) One aspect of the embodiment One aspect of the present embodiment will be described. FIG. 4 is an explanatory diagram of one aspect of the present embodiment.

As shown in FIG. 4, the client device 10 generates a request for estimating the amount of damage to a building damaged by a disaster (hereinafter referred to as "damaged building") in accordance with an instruction from the user, and sends the generated request. Send to server 30 . The user of the client device 10 is, for example, an owner of a damaged building, an insured person of damage insurance (for example, fire insurance or earthquake insurance) covering the damaged building, or an employee of an insurance company, a public institution, or a construction company. is.

Server 30 identifies the damaged building based on the request. The server 30 acquires the three-dimensional survey result TD1 corresponding to the specified damaged building. Here, the three-dimensional survey result TD1 is the result of the three-dimensional survey performed on the damaged building after the occurrence of the disaster. The three-dimensional survey result TD1 is represented, for example, as three-dimensional point cloud data. Information about the three-dimensional survey result TD1 (for example, the three-dimensional survey result TD1 itself or information indicating its location) may be included in the request from the client device 10 .

3D surveying is conducted by public institutions, insurance companies, construction companies, surveying companies, non-life insurance appraisers, owners of damaged buildings, or insured persons of non-life insurance covering damaged buildings (hereinafter referred to as "surveyor"). ) may be performed. For example, after the occurrence of a disaster, the surveyors may perform three-dimensional surveys all at once for areas damaged by the disaster (hereinafter referred to as "stricken areas"), or individually perform three-dimensional surveys for damaged buildings. may be performed as necessary. Three-dimensional surveying can be performed by a depth sensor mounted on at least one of an unmanned aerial vehicle, a mobile terminal, or a vehicle. The depth sensor is, for example, Lidar (Light Detection and Ranging), but is not limited to this.

The server 30 applies the estimation model EM1 to the input data based on the acquired three-dimensional survey result TD1 to estimate the amount of damage to the damaged building. The estimation model EM1 is constructed using machine learning about the relationship between the appearance of multiple buildings that have suffered damage in the past from disasters after the disaster (e.g., 3D survey results) and the actual amount of damage. be done.

The server 30 generates a response regarding the estimated damage amount and transmits the response to the client device 10 .

In this way, the information processing system 1 estimates the amount of damage that a disaster-stricken building has suffered due to a disaster, for example, without the need for man-hour estimation by humans. In other words, the amount of damage can be estimated at low cost and in a short period of time, and the restoration work of the damaged building can be made efficient.

(3) Database The database of this embodiment will be described. The following databases are stored in the storage device 31.

(3-1) Contract Database The contract database of this embodiment will be described. FIG. 5 is a diagram showing the data structure of the contract database of this embodiment.

The contract database stores contract information. The contract information is information related to insurance (in particular, non-life insurance such as fire insurance or earthquake insurance).
As shown in FIG. 5, the contract database includes a "contract ID" field, a "contractor" field, an "insured person" field, and an "insured object" field. Each field is associated with each other.

A contract ID is stored in the "contract ID" field. The contract ID is information for identifying contract information. The value of the contract ID may be the same as or different from the policy number of the insurance contract corresponding to the contract information. The contract database may further include fields for storing information about the policy number, if the contract ID value is different from the policy number.

The "contractor" field stores contractor information. The policyholder information is information about the policyholder of the insurance contract identified by the contract ID. By way of example, subscriber information may include, for example, information regarding the subscriber's name (eg, name or trade name), date of birth, address, or contact information.

The "insured person" field stores insured person information. The insured person information is information about the insured person of the insurance contract specified by the contract ID. As an example, the insured person information may include, for example, information regarding the insured person's name, date of birth, address, or contact information.

The "insurance coverage" field stores insurance coverage information. The insurance target information is information related to the target of the insurance contract (for example, building) specified by the contract ID.
In the example of FIG. 5, the coverage field includes a "Location" subfield, a "Usage" subfield, a "Structure" subfield, and a "Coverage" subfield.

The "location" subfield stores location information. The location information is information (eg, address information or geographical coordinate information) regarding the location of the insurance policy object identified by the policy ID.

The "Use" subfield stores use information. The use information is information on the use of the insurance contract object specified by the contract ID. As an example, the use of a building is classified into one of "dedicated housing", "apartment housing", "combined housing", or "other buildings".

The "structure" subfield stores structure information. Structural information is information related to the structural classification of the insurance contract object identified by the contract ID. As an example, when the object of insurance is a residential property, the structure class is classified into one of M structure (condominium structure), T structure (fire-resistant structure), or H structure (non-fire-resistant structure).

The "insurance money" subfield stores insurance money information. The insurance money information is information about the insurance money set for the insurance contract specified by the contract ID.

The insurance information may further include information not shown in FIG. 5, such as information regarding the terms of the insurance contract.
As a first example, insurance information may include valuation information. The valuation method information is information relating to the method of calculating the valuation of the insurance contract specified by the contract ID. For example, if the value of the valuation method information corresponds to "replacement value", the valuation of the insured object shall be of the same structure, quality, use, scale, type, and ability as the insured object at the place of occurrence at the time of damage occurrence. determined by the amount required to rebuild or reacquire the When the value of the valuation method information is "current value", the valuation of the insured object is determined by the amount obtained by subtracting the amount of wear and tear due to use from the replacement cost.
As a second example, insurance information may include contract term information. The contract period information is information on the expiration date of the insurance contract identified by the contract ID.
As a third example, insurance information may include premium information. The premium information is information related to the premium of the insurance contract specified by the contract ID. For example, the premium information may include the amount of the premium or information about how the premium is paid.

(4) Information processing Information processing according to the present embodiment will be described. FIG. 6 is a flow chart of information processing in this embodiment. FIG. 7 is a diagram showing an example of a screen displayed during information processing according to this embodiment. FIG. 8 is a diagram showing an example of a screen displayed during information processing according to this embodiment. FIG. 9 is a diagram showing an example of a screen displayed during information processing according to this embodiment.

For example, the information processing in FIG. Start.

As shown in FIG. 6, the client device 10 accepts a user instruction (S110).
Specifically, the client device 10 displays the screen P10 in FIG. The screen P10 corresponds to a UI (User Interface) that receives various user instructions regarding requests for automatic review of damage insurance for damaged buildings. The screen P10 includes field objects F10a to F10b and operation objects B10a to B10c.

The field object F10a accepts the input of the insurance policy number for the damaged building.
The field object F10b accepts input of the location of the damaged building.
The user selects one of the field objects F10a-F10b and inputs information. Based on this information, the server 30 identifies the damaged building.

The operation object B10a receives an operation for searching the storage device 11 for a file corresponding to the repair estimate for the damaged building (hereinafter referred to as "repair estimate file") and preparing to upload the file.
The operation object B10b generates a repair estimate file by photographing the repair estimate for the damaged building with a camera (not shown) provided on the client device 10, and performs an operation to prepare for uploading the file. accept.
The user selects one of the operation objects B10a-B10b to prepare for uploading the repair estimate file. When the preparation is completed, the display object A10 is arranged on the screen P10. The display object A10 displays an icon of a repair estimate file.
The operation object B10c accepts an operation for issuing a request.

The screen P10 is merely an example of a UI for receiving various user instructions regarding requests, and various modifications are possible. The screen P10 may be configured to further accept input of at least one of policyholder information or insured information for property insurance, information regarding a disaster summary, or information regarding a damage summary. The screen P10 is configured to accept a file of documents (for example, an insurance claim form, a disaster certificate, etc.) required for applying for damage insurance, in addition to the repair estimate, or information on items to be described in the documents. good too. The screen P10 may be configured to receive information (for example, a numerical value) indicating the total repair estimate amount for the damaged building in addition to or instead of the repair estimate file.

After step S110, the client device 10 executes the request (S111).
Specifically, the client device 10 generates a request based on the user instruction received in step S110 and transmits the request to the server 30. FIG. As an example, when the operation object B10c is selected on the screen P10, the client device 10 displays the input contents of the field objects F10a to F10b (information that can identify the damaged building) and the result of the operation on the operation object B10a to B10b (repair Quote file) and generate a request.

After step S111, the server 30 specifies contract information (S130).
Specifically, the server 30 receives the request transmitted in step S111. The server 30 refers to the contract database (FIG. 5) and identifies contract information for damaged buildings. As a first example, the server 30 identifies contract information based on the policy number included in the request. As a second example, the server 30 identifies contract information based on location information included in the request.

After step S130, the server 30 acquires the three-dimensional survey result (S131).
Specifically, the server 30 acquires the three-dimensional survey result of the damaged building using the information included in the request. As a first example, the server 30 acquires the three-dimensional survey results of the damaged building from a database constructed in the storage device 31 or an external device. As a second example, the server 30 acquires the three-dimensional survey results of the disaster area from a database built in the storage device 31 or an external device, and extracts a part of the three-dimensional survey results based on the location information of the damaged building. By doing so, the three-dimensional survey result of the damaged building is acquired. As a third example, the server 30 retrieves the three-dimensional survey results included in the request.

After step S131, the server 30 estimates the amount of damage (S132).
Specifically, the server 30 applies an estimation model to the input data based on the three-dimensional survey results obtained in step S131, thereby estimating the amount of damage to the damaged building. Here, the input data may be configured to be based on the results of three-dimensional surveying performed before the occurrence of the disaster, in addition to the results of three-dimensional surveying performed after the occurrence of the disaster. . Alternatively, the input data may be configured to be based on the results of three-dimensional surveying of the damaged building after the occurrence of the disaster, but not based on the results of three-dimensional surveying performed before the occurrence of the disaster. .

The estimation model is associated with input data based on the results of three-dimensional surveying of each building that has been damaged by a disaster in the past, and the input data, and A trained model created by supervised learning using a learning data set containing correct data that can identify the actual amount of damage suffered by each building in the disaster, or a derived model or distilled model of the trained model corresponds to Also, the estimation model may be constructed for each use or structure of the building, and used accordingly.

Here, the amount of damage includes the cost necessary to restore the damaged building to the state before the disaster (hereinafter referred to as "restoration cost"). The amount of damage may be only the restoration cost, or may be the sum of the restoration cost and the incidental cost. Ancillary costs are costs associated with restoration, and can include, for example, first aid costs, cause investigation costs, and the like. Alternatively, the amount of damage may be estimated separately for restoration costs and incidental costs.

After step S132, the server 30 performs a validity check (S133).
Specifically, the server 30 acquires information indicating the estimated total repair cost for the damaged building based on the information included in the request. As a first example, the server 30 extracts information indicating the repair estimate total from the repair estimate file included in the request. As a second example, the server 30 retrieves information indicating the total repair estimate included in the request.

The server 30 determines whether or not the estimated error condition is satisfied based on the total estimated repair cost and the amount of damage estimated in step S132. Estimated error conditions can include, for example, at least one of the following.
・The absolute value of the difference between the estimated total repair cost and the damage amount exceeds the first threshold (≧0) ・The ratio of the total estimated repair cost to the damage amount exceeds the second threshold value (≧1) ・The ratio of the damage amount to the total estimated repair cost ratio exceeds a third threshold (≥1)

If at least one estimate error condition is satisfied, the server 30 notifies that the estimated damage amount or repair estimate may not be appropriate, and terminates the information processing in FIG. If the estimated amount of damage is excessively large with respect to the total repair estimate, the estimated amount of damage may not be appropriate (the damage may be overestimated). If the estimated damage amount is excessively small relative to the total estimated repair cost, the estimated damage amount may be invalid (underestimating the damage) or the repair estimate may be overcharged. A contact may include, for example, at least one of a user, an insurance company, a construction company, or a public authority.

As an example, in response to the notification from the server 30, the client device 10 displays the screen P11 of FIG. The screen P11 includes a display object A11. Display object A11 displays an error message. The error message contains text that tells you that the automatic audit did not end normally, text that tells you that an on-site inspection by an expert witness is required, and information that prompts you to make an appointment for an on-site inspection (e.g., phone number and information on the booking page). link).

After step S133, the server 30 performs compensation amount assessment (S134).
Specifically, server 30 assesses the amount of compensation for the damaged building based on the contract information identified in step S130 and the amount of damage estimated in step S132. As an example, the server 30 assesses the compensation amount so that it does not exceed either the insurance amount specified by the contract information or the damage amount. The server 30 may assess the compensation amount as, for example, the full estimated repair amount, or the total estimated repair amount multiplied by a predetermined percentage less than one.

After step S134, the server 30 executes a response (S135).
Specifically, the server 30 generates a response including information regarding the amount of compensation assessed in step S134 and transmits the response to the client device 10 .

After step S135, the client device 10 performs screen display (S112).
Specifically, the client device 10 receives the response sent in step S135. The client device 10 displays a screen based on the response on the display 21. FIG. As an example, the client device 10 displays the screen P12 of FIG. 9 on the display 21. FIG. The screen P12 includes a display object A12. The display object A12 displays the result of automatic examination. The results of the automatic review include a statement that the automatic review was completed successfully, information regarding the assessed amount (ie, the amount of compensation), and information indicating when payment will be made.

(5) Summary As described above, the server 30 of this embodiment acquires the results of the three-dimensional survey performed on the damaged building after the occurrence of the disaster, and inputs data based on the results of the three-dimensional survey. By applying the estimation model to , the amount of damage to the affected building is estimated. As a result, the amount of damage to a damaged building can be estimated at low cost and in a short period of time, and the restoration work of the damaged building can be made more efficient.

The server 30 may specify the contract information of the damage insurance covering the damaged building, and assess the amount of compensation for the damaged building based on the contract information and the estimated amount of damage. As an example, the server 30 may assess the compensation amount so as not to exceed either the insurance amount specified by this contract information or the estimated damage amount. This makes it possible to provide the user with an automatic examination service for casualty insurance.

The server 30 acquires information indicating the total estimated repair amount for the damaged building, and if the absolute value of the difference between the estimated damage amount and the estimated total repair amount exceeds the first threshold, the repair for the damage amount is performed. If the ratio of the estimated total amount exceeds a second threshold, the notification may be provided if the ratio of the damage amount to the estimated repair amount exceeds a third threshold. This can inform the parties that the damage estimate may not be reasonable (understate or overestimate the damage) or that the repair estimate may be overcharged.

The estimation model is associated with input data based on the results of three-dimensional surveying of each building that has been damaged by a disaster in the past, and with the input data, and A trained model created by supervised learning using a training data set containing correct data that can identify the actual amount of damage suffered by each building in the disaster, or a derived model or distilled model of the trained model may be equivalent to This makes it possible to statistically estimate the actual amount of damage to the damaged building.

The damage amount may be defined to include restoration costs. Damage may also be defined to include incidental costs in addition to restoration costs. This makes it possible to estimate the amount of damage suitable for automatic examination of casualty insurance.

The three-dimensional survey may be performed by a depth sensor mounted on at least one of an unmanned aerial vehicle, a mobile terminal, or a vehicle. As a result, even if the surveyor does not have specialized knowledge or skills, it is possible to obtain three-dimensional survey results that can be used to estimate the amount of damage.

The input data for the estimation model may be configured to be based on the results of three-dimensional surveying of the damaged building after the occurrence of the disaster and not based on the results of three-dimensional surveying before the occurrence of the disaster. good. As a result, the amount of damage can be estimated even for a damaged building for which three-dimensional surveying has not been performed before the occurrence of the disaster.

(6) Modification A modification of the present embodiment will be described.

(6-1) Modification 1
Modification 1 will be described. Modification 1 is an example of evaluating the total damage in the disaster area. FIG. 10 is a flowchart of information processing according to Modification 1. As shown in FIG.

The information processing in FIG. 10 starts when the server 30 receives a request from the client device 10, for example.

As shown in FIG. 10, the server 30 acquires the three-dimensional survey result (S230).
Specifically, the server 30 acquires the three-dimensional survey result of the disaster area. As an example, the server 30 acquires the three-dimensional survey results of the disaster area from a database constructed in the storage device 31 or an external device.

After step S230, the server 30 associates the object with the survey result (S231).
Specifically, server 30 associates a subset of the three-dimensional survey results acquired in step S230 with each of the plurality of objects existing in the disaster area. Here, the object is land that exists in the disaster area, or a building or structure attached to the land. Land can include any kind of land such as, for example, residential land, rice fields, fields, forests, rivers, and the like. The server 30 acquires the three-dimensional survey result for each object by performing image processing for classifying the three-dimensional survey result into, for example, land, building, or construction. Alternatively, the server 30 may obtain the three-dimensional survey results for each object by dividing the three-dimensional survey result of the disaster area based on the location information of multiple objects existing in the disaster area.

After step S230, the server 30 performs an estimation (S232) of the amount of damage for each object.
Specifically, the server 30 applies the estimation model to the input data based on the three-dimensional survey results for each object acquired in step S230, thereby estimating the amount of damage to the object. Here, the input data may be based on the results of three-dimensional surveying performed before the occurrence of the disaster in addition to the results of three-dimensional surveying performed after the occurrence of the disaster. good. Alternatively, the input data may be configured to be based on the results of three-dimensional surveying of the object after the occurrence of the disaster, but not based on the results of three-dimensional surveying performed before the occurrence of the disaster. .

The estimation model consists of input data based on the results of three-dimensional surveying of each target object that has been damaged by a disaster in the past, and the input data associated with each target object. , and a learned model created by supervised learning using a training data set containing correct data that can identify the actual amount of damage suffered by each object due to the disaster, or a derived model of the learned model Or it corresponds to the distillation model. An estimation model may be constructed for each type of object and used properly.

Here, the amount of damage includes the restoration cost of the object. The amount of damage may be only the restoration cost, or may be the sum of the restoration cost and the incidental cost. Alternatively, the amount of damage may be estimated separately for restoration costs and incidental costs.

After step S232, the server 30 evaluates the total amount of damage (S233).
Specifically, the server 30 evaluates the total amount of damage in the disaster area based on the amount of damage estimated in step S232 for each of the plurality of objects. As an example, the server 30 calculates the total amount of damage by adding up the amounts of damage over a plurality of objects. The server 30 may send a response to the client device 10 including information regarding the total damage amount.

As described above, the server 30 of Modification 1 obtains the results of three-dimensional surveying of each of a plurality of objects existing in the disaster area after the occurrence of the disaster, and obtains the results of the three-dimensional surveying. Estimate the amount of damage for each of the objects in question by applying an estimation model to the input data based on Furthermore, the server 30 evaluates the total amount of damage in the disaster area based on the amount of damage estimated for each object. This makes it possible to estimate the total amount of damage in a disaster area at low cost and in a short period of time, and to provide useful information, for example, for planning a public subsidy system.

(7) Other Modifications The storage device 11 may be connected to the client device 10 via the network NW. The display 21 may be attached to the client device 10 or may be attached externally. Storage device 31 may be connected to server 30 via network NW.

Each step of the information processing described above can be executed by either the client device 10 or the server 30 . Also, in the above description, an example of executing each step in each process in a specific order has been shown, but the execution order of each step is not limited to the example described as long as there is no dependency. Furthermore, in the above description, an example in which the information processing system of the embodiment is implemented by a client/server system has been shown. However, the information processing system of the embodiments can also be implemented by a stand-alone computer.

In this embodiment, the server 30 identifies contract information for damaged buildings based on a request from the client device 10 . However, the server 30 acquires the three-dimensional survey result of the damaged building without depending on the request from the client device 10, and identifies the location of the damaged building based on the geographical coordinates associated with the three-dimensional survey result. may The server 30 may identify the contract information of the damaged building by extracting the contract information corresponding to the identified location from the contract database. As a result, without waiting for an application from a user, it is possible to automatically examine damage insurance for a damaged building. For example, an insurance company can estimate how much payment will be required in a disaster area before accepting applications from individual users in the disaster area.

In the present embodiment, an example of providing an automatic examination of casualty insurance for damaged buildings has been shown. However, instead of such automatic examination, it is also possible to provide a service for estimating the amount of damage to damaged buildings. In this case, steps S130 and S133 to S134 in FIG. 6 can be omitted.

(8) Supplementary notes The matters explained in the embodiment and the modified example are additionally noted below.

(Appendix 1)
a computer (30);
Means for acquiring the results of three-dimensional surveying performed after the occurrence of the disaster for the damaged building damaged by the disaster (S131);
Based on the results of three-dimensional surveying, an estimation model built using machine learning regarding the relationship between the post-disaster appearance of multiple buildings that have suffered damage in the past and the actual amount of damage. means for estimating the amount of damage to the damaged building by applying it to the input data (S132);
A program that acts as a

(Appendix 2)
the computer,
Means for specifying contract information of non-life insurance covering the damaged building (S130);
means for assessing the amount of compensation for the damaged building based on the contract information of the property insurance covering the damaged building and the amount of damage to the damaged building (S134);
1. The program of claim 1, further functioning as a

(Appendix 3)
The means for assessing the compensation amount assesses the compensation amount so as not to exceed either the insurance money specified by the contract information of the property insurance for the damaged building and the estimated damage amount of the damaged building.
A program according to Appendix 2.

(Appendix 4)
The means for identifying the contract information identifies the location of the damaged building based on the geographical coordinate information associated with the results of the three-dimensional survey, and the contract information corresponding to the location in the contract database in which non-life insurance contract information is registered. to extract the
A program according to Appendix 2 or Appendix 3.

(Appendix 5)
the computer,
Means for acquiring information indicating the total estimated repair cost for the damaged building (S130);
If the absolute value of the difference between the total estimated repair amount and the amount of damage estimated for the damaged building exceeds the first threshold, if the ratio of the total estimated repair amount to the amount of damage exceeds the second threshold, means for notifying when the damage amount ratio exceeds the third threshold (S133);
5. The program according to any one of appendices 1 to 4, further functioning as

(Appendix 6)
The estimation model is associated with input data based on the results of three-dimensional surveying of each building that has been damaged by a disaster in the past, and the input data, and A trained model created by supervised learning using a learning data set containing correct data that can identify the actual amount of damage suffered by each building in the disaster, or a derived model or distilled model of the trained model Equivalent to
The program according to any one of Appendices 1 to 5.

(Appendix 7)
The amount of damage to a damaged building includes the costs necessary to restore the damaged building to its pre-disaster state.
The program according to any one of Appendices 1 to 6.

(Appendix 8)
The amount of damage to the damaged building further includes incidental costs for restoring the damaged building to its pre-disaster state.
The program according to any one of Appendices 1 to 7.

(Appendix 9)
Three-dimensional surveying of the damaged building is performed by a depth sensor mounted on at least one of an unmanned aircraft, a mobile terminal, or a vehicle.
The program according to any one of Appendices 1 to 8.

(Appendix 10)
The input data is based on the results of three-dimensional surveying of the damaged buildings after the occurrence of the disaster, and not based on the results of three-dimensional surveying of the damaged buildings before the occurrence of the disaster. 9. A program according to any of Appendix 9.

(Appendix 11)
the computer,
Acquire the results of three-dimensional surveys performed after the occurrence of the disaster for each of multiple objects that are buildings or structures attached to the land existing in the disaster area that suffered damage due to the disaster. Means (S230),
Constructed using machine learning about the relationship between multiple lands damaged by disasters in the past, or the appearance of buildings or structures attached to the lands after the occurrence of the disaster and the actual amount of the damage. means for estimating the amount of damage for each of a plurality of objects present in the disaster area by applying the estimation model to the input data based on the results of the three-dimensional survey for each of the objects (S232);
Means for evaluating the total amount of damage in the disaster area based on the amount of damage estimated for each of the plurality of objects (S233);
A program that functions as

(Appendix 12)
a means for acquiring the results of a three-dimensional survey performed after the occurrence of the disaster for the building damaged by the disaster (S131);
Based on the results of three-dimensional surveying, an estimation model built using machine learning regarding the relationship between the post-disaster appearance of multiple buildings that have suffered damage in the past and the actual amount of damage. means for estimating the amount of damage to the damaged building by applying it to the input data (S132);
An information processing device (30) comprising:

(Appendix 13)
a computer (30)
a step of acquiring the results of a three-dimensional survey performed after the occurrence of the disaster for the damaged building damaged by the disaster (S131);
Based on the results of three-dimensional surveying, an estimation model built using machine learning regarding the relationship between the post-disaster appearance of multiple buildings that have suffered damage in the past and the actual amount of damage. A step of estimating the amount of damage to the damaged building by applying it to the input data (S132);
a method comprising

(Appendix 14)
comprising a client device (10) and a server (30),
The client device
means (S111) for generating a request for estimating the amount of damage to a building damaged by a disaster in response to an instruction from a user;
means for transmitting the request to the server (S111);
The server
means for obtaining a request (S130);
means for obtaining, upon request, the results of a three-dimensional survey performed on the damaged building after the occurrence of the disaster (S131);
Based on the results of three-dimensional surveying, an estimation model built using machine learning regarding the relationship between the post-disaster appearance of multiple buildings that have suffered damage in the past and the actual amount of damage. means for estimating the amount of damage to the damaged building by applying it to the input data (S132);
means for generating a response regarding the estimation result of the damage amount of the damaged building (S135);
and means for sending a response to the client device (S135).

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited to the above embodiments. Also, the above embodiments can be modified and modified in various ways without departing from the gist of the present invention. Also, the above embodiments and modifications can be combined.

1: information processing system 10: client device 11: storage device 12: processor 13: input/output interface 14: communication interface 21: display 30: server 31: storage device 32: processor 33: input/output interface 34: communication interface

Claims (14)

the computer,
Means for acquiring the results of three-dimensional surveying performed after the occurrence of the disaster for the damaged building damaged by the disaster;
An estimation model built using machine learning regarding the relationship between the appearance of multiple buildings that have suffered damage in the past from the disaster after the disaster and the actual amount of the damage is applied to the results of the three-dimensional survey. means for estimating the amount of damage to the damaged building by applying it to the input data based on
A program that acts as a said computer,
means for specifying contract information of non-life insurance covering the damaged building;
Means for assessing the amount of compensation for the damaged building based on the contract information of the non-life insurance covering the damaged building and the amount of damage to the damaged building;
2. The program of claim 1, further functioning as a The means for assessing the compensation amount assesses the compensation amount so as not to exceed either the insurance money specified by the contract information of the damage insurance for the damaged building and the estimated damage amount of the damaged building. do,
3. A program according to claim 2. The means for identifying the contract information identifies the location of the damaged building based on the geographical coordinate information associated with the result of the three-dimensional survey, and corresponds to the location in a contract database in which non-life insurance contract information is registered. extract contract information to
4. The program according to claim 2 or 3. said computer,
means for acquiring information indicating a total estimated repair cost for the damaged building;
If the absolute value of the difference between the total estimated repair amount and the damage amount estimated for the damaged building exceeds a first threshold, and if the ratio of the total estimated repair amount to the damage amount exceeds a second threshold, the total estimated repair amount means to notify when the ratio of the amount of damage to the third threshold exceeds the third threshold,
5. The program according to any one of claims 1 to 4, further functioning as a program. The estimation model includes input data based on the results of three-dimensional surveying of each building that has been damaged by a disaster in the past after the occurrence of the disaster, and associated with the input data, A trained model created by supervised learning using a learning data set containing correct data that can identify the actual amount of damage suffered by each building due to the disaster, or a derived model or distillation of the trained model Equivalent to the model,
6. The program according to any one of claims 1 to 5. The damage amount of the damaged building includes the cost necessary to restore the damaged building to the state before the disaster occurred,
A program according to any one of claims 1 to 6. The amount of damage to the damaged building further includes incidental costs for restoring the damaged building to its pre-disaster state.
A program according to any one of claims 1 to 7. Three-dimensional surveying of the damaged building is performed by a depth sensor mounted on at least one of an unmanned aircraft, a mobile terminal, or a vehicle.
A program according to any one of claims 1 to 8. The input data is based on the results of three-dimensional surveying of the damaged building after the occurrence of the disaster, and is not based on the results of three-dimensional surveying of the damaged building before the occurrence of the disaster. A program according to any one of claims 1 to 9. the computer,
Acquire the results of three-dimensional surveys performed after the occurrence of the disaster for each of multiple objects that are buildings or structures attached to the land existing in the disaster area that suffered damage due to the disaster. means,
Constructed using machine learning about the relationship between multiple lands damaged by disasters in the past, or the appearance of buildings or structures attached to the lands after the occurrence of the disaster and the actual amount of the damage. Means for estimating the amount of damage to each of the plurality of objects by applying the estimation model to the input data based on the results of the three-dimensional survey for each of the plurality of objects existing in the disaster area;
Means for evaluating the total amount of damage in the disaster area based on the amount of damage estimated for each of the plurality of objects;
A program that functions as a means for acquiring the results of a three-dimensional survey performed after the occurrence of a disaster on a building damaged by a disaster;
An estimation model built using machine learning regarding the relationship between the appearance of multiple buildings that have suffered damage in the past from the disaster after the disaster and the actual amount of the damage is applied to the results of the three-dimensional survey. means for estimating the amount of damage to the damaged building by applying it to input data based on
An information processing device comprising: the computer
a step of acquiring the results of a three-dimensional survey performed after the occurrence of the disaster for a building damaged by the disaster;
An estimation model built using machine learning regarding the relationship between the appearance of multiple buildings that have suffered damage in the past from the disaster after the disaster and the actual amount of the damage is applied to the results of the three-dimensional survey. estimating the amount of damage to the damaged building by applying to input data based on
An information processing method comprising: comprising a client device and a server;
The client device
means for generating, in response to an instruction by a user, a request for an estimate of the amount of damage to a building damaged by a disaster;
means for sending said request to said server;
The server is
means for obtaining the request;
means for obtaining, in response to the request, a result of a three-dimensional survey performed on the damaged building after the occurrence of the disaster;
An estimation model built using machine learning regarding the relationship between the appearance of multiple buildings that have suffered damage in the past from the disaster after the disaster and the actual amount of the damage is applied to the results of the three-dimensional survey. means for estimating the amount of damage to the damaged building by applying it to input data based on
means for generating a response regarding estimated damage amount of the damaged building;
and means for transmitting said response to said client device.

Images