Giant disaster service processing method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of huge disaster data processing, is applied to the technical field of digital finance, and particularly relates to a huge disaster business processing method and device, computer equipment and a storage medium. According to the invention, the huge disaster risk level of the huge disaster area is predicted through the huge disaster risk prediction model, and the compensation capability level corresponding to the reinsurance company is predicted through the compensation capability prediction model, so that the target company is determined from the reinsurance companies based on the huge disaster risk level and the compensation capability level, and the reinsurance contract is signed with the target company. Compared with manual evaluation for selecting the target company, the risk matching efficiency can be effectively improved.

Description

Technical Field

[0001] This invention relates to the field of disaster data processing technology, and is applied to the field of digital financial technology. In particular, it relates to a disaster business processing method, apparatus, computer equipment, and storage medium. Background Technology

[0002] In the field of digital finance, original insurance companies choose reinsurance companies mainly to mitigate excess losses from catastrophic events and emerging risks, and to enhance their solvency by leveraging the capital strength of reinsurance companies. At the same time, they optimize their capital structure through reinsurance, release funds for core business, and improve their overall risk resistance and sustainable development capabilities.

[0003] Currently, primary insurance companies typically select reinsurance companies through manual assessment. However, manual assessment relies heavily on past cooperation experience and industry reputation, which has significant subjectivity and limitations. This makes it difficult for primary insurance companies to accurately assess the reinsurance company's ability to bear catastrophic risks and emerging risks. Therefore, primary insurance companies face the problem of low risk matching efficiency when selecting reinsurance formulas. Summary of the Invention

[0004] This invention provides a method, apparatus, computer equipment, and storage medium for handling catastrophe business, in order to solve the problem of low risk matching efficiency in the existing technology of selecting reinsurance companies through manual assessment.

[0005] A disaster recovery service method includes: Obtain regional meteorological data for the disaster area, customer exposure data of the original insurance company in the disaster area, and claims settlement capacity data of the reinsurance company; The customer exposure data and the regional meteorological data are input into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area. The catastrophic risk level includes the corresponding catastrophic loss amount. The claim-paying capacity data of each reinsurance company is input into the claim-paying capacity prediction model to obtain the claim-paying capacity level corresponding to each reinsurance company. The claim-paying capacity level includes the corresponding catastrophe payout amount. Based on the amount of catastrophe loss and the amount of catastrophe compensation, a target company is determined from the reinsurance companies, and the target company is used to enter into a reinsurance contract with the original insurance company.

[0006] The above-mentioned catastrophe business handling methods may optionally include the corresponding reinsurance sharing ratio for the catastrophe risk level; After determining the target company from the reinsurance companies based on the aforementioned catastrophe risk level and the aforementioned payability level, the process further includes: Using blockchain technology, a smart contract is generated between the original insurance company and the target company based on the reinsurance sharing ratio; When a catastrophic loss occurs in the catastrophic area, the actual compensation amount for the catastrophic area is calculated so that the original insurance company and the target company can pay the actual compensation amount in accordance with the smart contract.

[0007] Optionally, after calculating the catastrophic loss amount in the catastrophic area so that the original insurance company and the target company can pay the catastrophic loss amount according to the smart contract, the above-mentioned catastrophic business processing method further includes: The amount of catastrophic loss, the amount of catastrophic loss, and the amount of actual compensation are input into the catastrophic compensation process audit model to obtain the compensation process audit result. The compensation process audit result is used to reflect the compliance of the compensation process and the rationality of the compensation result.

[0008] Optionally, the above-mentioned catastrophic business processing method may involve inputting the customer exposure data and the regional meteorological data into a catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area, including: The regional meteorological data and the customer exposure data are preprocessed respectively to extract regional meteorological features from the regional meteorological data and regional customer features from the customer exposure data. The regional meteorological characteristics and the regional customer characteristics are fused to obtain the regional risk characteristics; The regional risk characteristics are input into the catastrophic risk prediction model to obtain the catastrophic risk level.

[0009] Optionally, the above-mentioned disaster business processing method may preprocess the regional meteorological data and the customer exposure data separately to extract regional meteorological features from the regional meteorological data and regional customer features from the customer exposure data, including: The regional meteorological data is cleaned, and the cleaned regional meteorological data is converted into regional meteorological features through feature engineering. The customer exposure data is cleaned, and the cleaned customer exposure data is transformed into regional customer characteristics through feature engineering.

[0010] Optionally, the above-mentioned catastrophe business processing method may involve feature fusion of the regional meteorological characteristics and the regional customer characteristics to obtain regional risk characteristics, including: Based on the disaster-stricken area, spatial matching is performed on the meteorological characteristics of the area and the customer characteristics of the area to generate the risk characteristics of the area.

[0011] Optionally, the regional meteorological data in the above-mentioned disaster business processing method includes: wind speed information, rainfall information, and temperature and humidity information; The customer exposure data includes: geographic information, asset information, risk resistance capabilities, and business continuity. The compensation capacity data includes: solvency, compensation records, and actual assets.

[0012] A catastrophe business processing device, comprising: The data acquisition module is used to acquire regional meteorological data of the disaster area, customer exposure data of the original insurance company in the disaster area, and reimbursement capacity data of the reinsurance company. The risk prediction module is used to input the customer exposure data and the regional meteorological data into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area, wherein the catastrophic risk level includes the corresponding catastrophic loss amount; The capacity prediction module is used to input the claim-paying capacity data of each reinsurance company into the claim-paying capacity prediction model to obtain the claim-paying capacity level of each reinsurance company, wherein the claim-paying capacity level includes the corresponding catastrophe claim amount. The company matching module is used to determine a target company from the reinsurance companies based on the amount of catastrophic loss and the amount of catastrophic compensation. The target company is used to sign a reinsurance contract with the original insurance company.

[0013] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the aforementioned disaster recovery service processing methods.

[0014] A computer-readable storage medium storing a computer program that, when executed by a processor, implements any of the above-described catastrophe service processing methods.

[0015] This invention provides a method, apparatus, computer equipment, and storage medium for handling catastrophic events. It acquires regional meteorological data of the catastrophic area, customer exposure data of the original insurance company in the catastrophic area, and claims-paying capacity data of reinsurance companies. The customer exposure data and regional meteorological data are input into a catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area, which includes the corresponding catastrophic loss amount. The claims-paying capacity data of each reinsurance company are input into a claims-paying capacity prediction model to obtain the claims-paying capacity level of each reinsurance company, which includes the corresponding catastrophic compensation amount. Based on the catastrophic loss amount and the catastrophic compensation amount, a target company is determined from among the reinsurance companies. This target company is used to sign a reinsurance contract with the original insurance company. Therefore, this invention predicts the catastrophic risk level of the catastrophic area using a catastrophic risk prediction model and the corresponding claims-paying capacity level of the reinsurance companies using a claims-paying capacity prediction model. Based on the catastrophic risk level and claims-paying capacity level, a target company is determined from among the reinsurance companies, and a reinsurance contract is signed with the target company. Compared to manual assessment for selecting target companies, this method effectively improves risk matching efficiency. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an implementation of a disaster recovery service processing method provided in one embodiment of the present invention. Figure 2 This is a flowchart illustrating another implementation of a disaster recovery service processing method provided in one embodiment of the present invention. Figure 3 This is a flowchart illustrating another implementation of a disaster recovery service processing method provided in one embodiment of the present invention. Figure 4 This is a partial implementation flowchart of a disaster recovery service processing method provided in one embodiment of the present invention; Figure 5 This is a partial implementation flowchart of a disaster recovery service processing method provided in one embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a disaster recovery service processing device provided in one embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of a computer device provided in one embodiment of the present invention. Detailed Implementation

[0018] To make the technical problems solved, the technical solutions, and the beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0020] It should also be understood that the term “and / or” as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0021] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0022] Furthermore, in the description of this invention and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0023] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0024] This invention provides a method, apparatus, computer equipment, and storage medium for handling catastrophic events. It acquires regional meteorological data of the catastrophic area, customer exposure data of the original insurance company in the catastrophic area, and claims-paying capacity data of reinsurance companies. The customer exposure data and regional meteorological data are input into a catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area, which includes the corresponding catastrophic loss amount. The claims-paying capacity data of each reinsurance company are input into a claims-paying capacity prediction model to obtain the claims-paying capacity level of each reinsurance company, which includes the corresponding catastrophic compensation amount. Based on the catastrophic loss amount and the catastrophic compensation amount, a target company is determined from the reinsurance companies. This target company is used to sign a reinsurance contract with the original insurance company. Therefore, this invention predicts the catastrophic risk level of the catastrophic area using a catastrophic risk prediction model and the corresponding claims-paying capacity level of the reinsurance companies using a claims-paying capacity prediction model. Based on the catastrophic risk level and claims-paying capacity level, a target company is determined from the reinsurance companies, and a reinsurance contract is signed with the target company. Compared to manual assessment for selecting a target company, this method can effectively improve risk matching efficiency. Specific embodiments are described below.

[0025] In one embodiment, such as Figure 1 The diagram shown is a flowchart illustrating the implementation of a disaster recovery service processing method according to an embodiment of the present invention. This method is applicable to electronic devices with data processing capabilities, such as... Figure 1 As shown: S101: Obtain regional meteorological data of the disaster area, customer exposure data of the original insurance company in the disaster area, and claims settlement capacity data of the reinsurance company.

[0026] Regional meteorological data refers to dynamic monitoring data that reflects the environmental characteristics of disaster-prone areas. Specifically, it can be obtained by collecting wind speed, rainfall, and temperature and humidity information from meteorological stations, and is used to assess the probability of disaster occurrence and the intensity of its impact.

[0027] Customer exposure data refers to the asset distribution and risk tolerance data of the original insurance company's customers in the disaster area. For example, information on building locations, asset values ​​and disaster prevention facilities can be obtained through geographic information systems to quantify the economic losses that disasters may cause.

[0028] Claims capacity data refers to the financial status and historical claims records of reinsurance companies. For example, solvency adequacy ratio, current asset size, and past catastrophic claims cases can be extracted from publicly available financial statements to assess the upper limit of their risk-bearing capacity.

[0029] S102: Input customer exposure data and regional meteorological data into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area.

[0030] The catastrophic risk level includes the corresponding catastrophic loss amount.

[0031] In its specific implementation, the catastrophic risk prediction model in this embodiment includes, but is not limited to, any one of the following: random forest model and neural network model. Data features are extracted from customer exposure data and regional meteorological data, respectively, and then fused. The fused features are then input into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area. Specifically, the catastrophic risk prediction model establishes a mapping relationship between meteorological parameters and economic losses by training on historical disaster data and loss records. For example, when rainfall exceeds a threshold, the model automatically triggers flood disaster loss calculations.

[0032] S103: Input the claimability data of each reinsurance company into the claimability prediction model to obtain the claimability level of each reinsurance company. The claimability level includes the corresponding catastrophe payout amount.

[0033] In its implementation, the solvency prediction model in this embodiment includes, but is not limited to, any one of the KNN and SVM models. After extracting data features from the solvency data, these features are input into the catastrophe risk prediction model to obtain the catastrophe risk level of the catastrophe region. The solvency prediction model dynamically assesses indicators such as the financial liquidity and capital adequacy ratio of reinsurance companies. For example, when a company recently completes a large payout, the model automatically lowers its solvency level. Finally, through a numerical comparison of loss amounts and payout amounts, for example, prioritizing reinsurance companies whose payouts cover more than 120% of the loss amount, automated decision-making for risk diversification and capital optimization is achieved.

[0034] S104: Based on the amount of catastrophe loss and catastrophe compensation, a target company is selected from among the reinsurers. The target company is used to enter into a reinsurance contract with the original insurance company.

[0035] In a specific implementation, this embodiment can select the reinsurer corresponding to the catastrophe compensation amount that is greater than the amount of catastrophe loss as the target company, or select the reinsurer with the largest catastrophe compensation amount as the target company, so as to ensure that the target company has sufficient compensation capacity.

[0036] In summary, this application, through the fusion of meteorological and customer data, accurately quantifies the scope of disaster impact and the amount of asset loss. Simultaneously, the model-based solvency assessment avoids human bias, objectively reflecting the reinsurance company's actual solvency. The numerical matching mechanism directly achieves a quantitative connection between risk exposure and solvency, resolving the inefficiencies and inaccurate risk matching issues caused by the subjectivity of manual assessment. The joint analysis of meteorological and customer data enables the objective quantification of catastrophic risk, the dynamic assessment of the solvency model ensures the accuracy of reinsurance company selection, and the numerical matching mechanism shortens the traditionally weeks-long manual process to a few hours.

[0037] In one embodiment, such as Figure 2 As shown, the catastrophe risk level includes the corresponding reinsurance sharing ratio. Following step S104 in the above embodiment, the following steps may also be included: S201: Using blockchain technology, a smart contract is generated between the original insurance company and the target company based on the reinsurance contribution ratio.

[0038] S202: When a catastrophic loss occurs in a catastrophic area, calculate the actual compensation amount in the catastrophic area so that the original insurance company and the target company can pay the actual compensation amount in accordance with the smart contract.

[0039] In this embodiment, the blockchain technology refers to distributed ledger technology, which can be implemented using a consortium blockchain or a private blockchain architecture. A node consensus mechanism ensures data immutability, providing a trusted execution environment for smart contracts. Smart contracts are automated protocols written in code, specifically Ethereum smart contracts or Hyperledger Fabric chaincode, which automatically trigger the compensation process when preset conditions are met.

[0040] In one implementation, the reinsurance sharing ratio in this embodiment refers to the risk-sharing ratio agreed upon between the original insurance company and the reinsurance company. Specifically, a dynamic calculation model can be used to adjust the risk ratio according to the catastrophic risk level to ensure that the sharing ratio matches the risk level.

[0041] Specifically, after the catastrophe risk level is determined, the system writes the corresponding reinsurance sharing ratio into a blockchain smart contract, and the contract terms are solidified through encryption algorithms. When a loss occurs in a catastrophe-stricken area, the actual compensation amount is collected by IoT devices or third-party institutions and uploaded to the blockchain network, triggering the smart contract to automatically verify the data's authenticity. After successful verification, the contract code automatically executes the fund transfer, and the original insurance company and the target company complete the compensation according to the sharing ratio. In this process, the distributed storage characteristics of blockchain ensure that the contract content is transparent and verifiable, the consensus mechanism avoids the risk of single point of tampering, and the automated execution of the smart contract reduces the need for manual review.

[0042] In summary, this application, through the combination of blockchain and smart contracts, solidifies the sharing ratio in code, and the compensation process is automatically triggered by preset conditions, eliminating delays and errors caused by manual intervention. Simultaneously, data on-chain enables full traceability. This solves the problems of low efficiency and poor reliability of manual operations in the catastrophe compensation process, achieving transparency in the execution of compensation ratios and automation of contract execution, ensuring the auditability of compensation results, and reducing the risk of disputes arising from information asymmetry between the parties.

[0043] In one embodiment, such as Figure 3As shown, after step S202 in the above embodiment, this embodiment may further include the following steps: S301: Input the amount of catastrophic loss, the amount of catastrophic loss, and the actual compensation amount into the catastrophic compensation process audit model to obtain the compensation process audit results.

[0044] The results of the compensation process review are used to reflect the compliance of the compensation process and the reasonableness of the compensation results.

[0045] In one embodiment, the catastrophe compensation process review model in this embodiment refers to a computational model that automates the review of the compensation process through a preset algorithm. Specifically, it can be implemented using a machine learning model or a rule engine to analyze the logical correlation between the amount of catastrophe loss and the actual compensation amount, thereby verifying whether the compensation process complies with the preset rules.

[0046] In one embodiment, the compliance of the compensation process in this embodiment refers to whether the compensation operation follows the steps and conditions agreed in the smart contract. Specifically, this can be achieved by comparing the actual compensation steps with the contract terms to ensure that the execution of each link complies with laws and industry norms.

[0047] In one embodiment, the reasonableness of the compensation result refers to whether the actual compensation amount matches the catastrophe risk level and the reinsurance sharing ratio. Specifically, it can be achieved by using the deviation threshold detection method, which calculates the difference between the actual compensation amount and the predicted loss amount to determine whether the compensation result is within a reasonable range.

[0048] Specifically, after the smart contract executes the compensation payment, the catastrophe loss amount is repeatedly input into the audit model to verify data consistency and eliminate audit biases caused by data errors. After the actual compensation amount and the catastrophe loss amount are simultaneously input into the model, the model analyzes the correlation between the two, such as comparing the contractually agreed-upon sharing ratio with the actual compensation ratio, to identify any instances of non-compliance with the agreement. Furthermore, the model uses a pre-set compliance rule base to check whether the compensation operation fully covers the steps required by the contract, such as whether key aspects like loss confirmation, liability allocation, and fund transfer have been completed. The final audit result includes a process compliance score and a compensation reasonableness score, providing data support for subsequent dispute resolution or process optimization.

[0049] In summary, this application achieves full-process data-driven review through an automated model, which not only reduces errors caused by human intervention but also processes massive amounts of compensation data in real time, improving review speed and coverage. At the same time, it can automatically identify violations in the compensation process and avoid disputes caused by subjective judgment by quantitatively analyzing the reasonableness of compensation amounts, thereby improving the credibility and execution efficiency of catastrophe compensation business.

[0050] In one embodiment, such as Figure 4 As shown, step S102 in the above embodiment can be implemented through the following steps: S401: Preprocess the regional meteorological data and customer exposure data separately to extract regional meteorological features from the regional meteorological data and regional customer features from the customer exposure data.

[0051] Preprocessing refers to the standardization of regional meteorological data and customer exposure data. Specifically, it can be achieved by methods such as missing value removal, unit unification, and outlier detection, in order to eliminate data noise and improve input quality.

[0052] Specifically, the first step is to standardize and clean the regional meteorological data, such as unifying the units for parameters like wind speed and rainfall to eliminate errors caused by differences in data collection. Simultaneously, outlier detection is performed on fields like geographic coordinates and asset values ​​in the customer exposure data, for example, using the quartile method to identify values ​​outside the reasonable range. The cleaned data is then transformed into structured features through feature engineering, such as converting continuous regional meteorological data into graded indices and mapping customer exposure data to risk exposure coefficients.

[0053] S402: The regional meteorological characteristics and regional customer characteristics are fused to obtain the regional risk characteristics.

[0054] Among them, regional risk characteristics refer to quantitative indicators that reflect the overall risk level. Specifically, they can be expressed in the form of multidimensional feature vectors and used as input parameters for prediction models.

[0055] Specifically, in this embodiment, the spatial matching algorithm correlates regional meteorological characteristics with regional customer characteristics to obtain regional risk characteristics. For example, it overlays meteorological parameters within the same geographic grid with customer asset distribution to generate a comprehensive risk feature vector that includes meteorological influencing factors and customer vulnerability indicators.

[0056] S403: Input the regional risk characteristics into the catastrophic risk prediction model to obtain the catastrophic risk level.

[0057] Specifically, the catastrophe risk prediction model in this embodiment includes, but is not limited to, any one of the following: gradient boosting decision tree model, neural network model, etc. The regional risk characteristics are input into the catastrophe risk prediction model, and the output is a quantified catastrophe risk level.

[0058] In summary, this application eliminates data noise through standardized preprocessing, quantifies risk factors through feature engineering, and enables collaborative analysis of cross-dimensional data through spatial matching, constructing a complete automated data processing chain. This solves the risk assessment bias problem caused by low data quality and lack of correlation analysis in manual assessments. The preprocessing process ensures the reliability of input data, and the feature fusion mechanism enhances the correlation analysis capability of different risk factors, enabling the model to make accurate predictions based on high-quality structured features, thereby improving the efficiency of catastrophic risk matching.

[0059] In one embodiment, such as Figure 5 As shown, step S401 in the above embodiment can be implemented through the following steps: S501: Perform data cleaning on regional meteorological data and transform the cleaned regional meteorological data into regional meteorological features through feature engineering.

[0060] S502: Perform data cleaning on customer exposure data and transform the cleaned customer exposure data into regional customer characteristics through feature engineering.

[0061] Data cleaning refers to the process of improving the quality of raw data (such as the aforementioned regional meteorological data and customer exposure data). Specifically, this can involve using interpolation or record deletion to remove missing values, standardizing units through unit conversion formulas, and using Z-score or IQR methods for outlier detection to eliminate data noise interference with model training. Feature engineering refers to the process of converting the cleaned data into features that the model can recognize. Specifically, this can involve using principal component analysis or statistical aggregation methods to extract temporal features from meteorological data, and using spatial clustering or geocoding to extract spatial distribution features from customer data, to construct structured features suitable for model input.

[0062] Specifically, parameters such as wind speed and rainfall in regional meteorological data may contain sensor errors or data omissions. Removing missing values ​​can prevent model bias caused by null values. Unit standardization converts Fahrenheit and Celsius temperature data from different sources into a unified dimension, eliminating feature weight imbalances. Outlier detection identifies typhoon wind speed records exceeding historical extremes, avoiding the influence of extreme noise on prediction results. In the feature engineering stage, the cleaned meteorological time-series data is converted into 24-hour moving average wind speed features, while customer-exposed data is aggregated by geographic grid to form regional asset density features, creating standardized inputs for model fusion.

[0063] In summary, this application, through source data cleaning and feature transformation, accurately depicts the spatial distribution characteristics of customer assets while preserving the dynamic changes of meteorological elements. This provides a highly consistent data foundation for subsequent risk level calculations and effectively solves the model prediction error problem caused by defects in the quality of the original data. The data cleaning process effectively eliminates the interference of missing and outlier values ​​on feature extraction, the unified unit operation solves the problem of inconsistent dimensions of multi-source data, and the feature engineering transformation constructs standardized features adapted to model processing. This enables the catastrophe risk prediction model to accurately assess regional risk levels based on high-quality input data, thereby improving the risk matching efficiency between primary insurance companies and reinsurance companies.

[0064] In one embodiment, step S402 in the above embodiment can be implemented as follows: based on the catastrophic area, spatial matching is performed on the regional meteorological characteristics and regional customer characteristics to generate regional risk characteristics.

[0065] Spatial matching refers to aligning data from different sources according to geographic spatial coordinates. Specifically, this can be achieved using geographic information system spatial overlay analysis technology. By unifying the geographic grid of regional meteorological data with the geographic boundaries of customer exposure data using a coordinate system, different data layers can form a precise correspondence in spatial dimensions.

[0066] Specifically, in the process of catastrophic risk assessment, regional meteorological data is typically stored in a kilometer-scale grid format, while customer exposure data is based on administrative regions or specific addresses. Spatial matching techniques are used to spatially correlate meteorological grid data with customer address coordinates. For example, a geographically weighted regression method is employed to calculate the meteorological risk value for each customer location by weighting the meteorological characteristics of its own grid and adjacent grids. Simultaneously, kernel density estimation is performed on customer asset distribution data to generate a spatial distribution map with the same resolution as the meteorological grid. This spatial feature fusion mechanism can eliminate risk assessment biases caused by differences in data spatial scale in traditional methods, such as avoiding the incorrect association of average rainfall at the county-level administrative region with street-level customer asset data.

[0067] In summary, this solution achieves a refined correspondence between the physical impact range of a disaster and the spatial distribution of customer assets through precise matching of geospatial coordinate systems. It resolves the risk assessment error caused by inconsistent data spatial resolution and eliminates feature correlation errors resulting from the mismatch in spatial dimensions between meteorological and customer data. This allows for coupled calculations of disaster intensity distribution and customer vulnerability distribution at the same spatial scale. This precise spatial matching mechanism significantly improves the spatial consistency of input data for catastrophic loss prediction models, thereby enhancing the accuracy of risk level classification and providing a reliable data foundation for reinsurance decisions.

[0068] In one embodiment, the regional meteorological data includes: wind speed information, rainfall information, and temperature and humidity information; the customer exposure data includes: geographic information, asset information, risk resistance capability, and business continuity; and the compensation capability data includes: solvency, compensation records, and actual assets.

[0069] Among these, wind speed information refers to wind intensity data collected by meteorological monitoring equipment in disaster-prone areas, specifically using meteorological satellite remote sensing or real-time monitoring data from ground meteorological stations, used to quantify the destructive power of meteorological disasters such as typhoons and hurricanes. Rainfall information refers to the total precipitation per unit time, specifically obtained using rain gauges or radar inversion technology, used to assess the risk of secondary disasters such as floods and mudslides. Temperature and humidity information refers to combined air temperature and relative humidity data, specifically collected using temperature and humidity sensor networks, used to predict extreme weather events such as heat waves and droughts. Geographic information refers to the latitude and longitude coordinates and topographic features of the client's location, specifically using geographic information systems for spatial annotation, used to determine the spatial distribution of risk-exposed areas. Asset information refers to the value of the client's fixed assets in the disaster-prone area, specifically extracted from the insured property registration database, used to calculate the potential economic loss scale. Risk resistance capability refers to the level of the client's disaster prevention facilities construction, specifically quantified using building structural safety assessment reports, used to correct the loss probability in risk prediction models. Business continuity refers to a reinsurance company's ability to maintain operations after a disaster. This can be measured by indicators of the completeness of the company's emergency response plan and is used to assess the duration of secondary losses. Solvency refers to the reinsurance company's capital-to-liability ratio, specifically using solvency adequacy ratio data disclosed by regulatory agencies to verify the reinsurance institution's financial soundness. Claims record refers to the reinsurance company's historical claims processing timeliness and achievement rate, specifically through statistical analysis of industry claims databases to assess its performance reliability. Actual assets refer to the size of a reinsurance company's realizable current assets, specifically using cash and cash equivalents data from financial statements to ensure cash flow support for catastrophe payouts.

[0070] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0071] In one embodiment, a disaster recovery service processing apparatus is provided, which corresponds one-to-one with the disaster recovery service processing methods described in the above embodiments. For example... Figure 6 As shown, this catastrophe business processing device includes a data acquisition module, a risk prediction module, a capacity prediction module, and a company matching module. Detailed descriptions of each functional module are as follows: The data acquisition module 601 is used to acquire regional meteorological data of the disaster area, customer exposure data of the original insurance company in the disaster area, and claims settlement capacity data of the reinsurance company; Risk prediction module 602 is used to input customer exposure data and regional meteorological data into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area. The catastrophic risk level includes the corresponding catastrophic loss amount. The capacity prediction module 603 is used to input the claim capacity data of each reinsurance company into the claim capacity prediction model to obtain the claim capacity level of each reinsurance company. The claim capacity level includes the corresponding catastrophe claim amount. The company matching module 604 is used to identify a target company from among the reinsurers based on the amount of catastrophe loss and catastrophe compensation. The target company is used to sign a reinsurance contract with the original insurance company.

[0072] In one embodiment, the catastrophe risk level includes a corresponding reinsurance sharing ratio. The catastrophe business processing device in this embodiment further includes a catastrophe claims module, used for: Using blockchain technology, a smart contract is generated between the original insurance company and the target company based on the reinsurance contribution ratio. When catastrophic losses occur in a catastrophic area, the actual compensation amount in the catastrophic area is calculated so that the original insurance company and the target company can pay the actual compensation amount in accordance with the smart contract.

[0073] In one embodiment, the catastrophe business processing device in this embodiment further includes a compensation review module, used for: Input the amount of catastrophic loss, the amount of catastrophic loss, and the actual amount of compensation into the catastrophic compensation process audit model to obtain the compensation process audit results. The compensation process audit results are used to reflect the compliance of the compensation process and the rationality of the compensation results.

[0074] In one embodiment, the risk prediction module 602 is used for: Regional meteorological data and customer exposure data are preprocessed separately to extract regional meteorological characteristics from regional meteorological data and regional customer characteristics from customer exposure data. Regional risk characteristics are obtained by fusing regional meteorological characteristics and regional customer characteristics. Regional risk characteristics are input into the catastrophic risk prediction model to obtain the catastrophic risk level.

[0075] In one embodiment, the risk prediction module 602 can be used for: The regional meteorological data is cleaned, and the cleaned regional meteorological data is transformed into regional meteorological features through feature engineering. We perform data cleaning on customer exposure data and then use feature engineering to transform the cleaned customer exposure data into regional customer characteristics.

[0076] In one embodiment, the risk prediction module 602 can be used for: Based on the disaster-prone areas, spatial matching of regional meteorological characteristics and regional customer characteristics is performed to generate regional risk characteristics.

[0077] In one embodiment, regional meteorological data includes: wind speed information, rainfall information, and temperature and humidity information; Customer-exposed data includes: geographic information, asset information, risk resistance capabilities, and business continuity. Claims-paying capacity data includes: solvency, claims history, and actual assets.

[0078] Specific limitations regarding the disaster recovery service processing device can be found in the limitations of the disaster recovery service processing method described above, and will not be repeated here. Each module in the aforementioned disaster recovery service processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0079] In one embodiment, a computer device is provided, such as Figure 7 As shown, it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the disaster recovery service processing method described in the above embodiments, for example... Figure 1 The disaster recovery service processing method shown, or Figures 2 to 5 As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the disaster recovery service device, for example... Figure 6 The functions of the disaster recovery service shown are not described again here to avoid duplication.

[0080] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the disaster recovery service processing method described in the above embodiments, for example... Figure 1 The disaster recovery service processing method shown, or Figures 2 to 5 As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the disaster recovery service device, for example... Figure 6 The functions of the disaster recovery service shown are not described again here to avoid duplication.

[0081] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for handling catastrophic events, characterized in that, include: Obtain regional meteorological data for the disaster area, customer exposure data of the original insurance company in the disaster area, and claims settlement capacity data of the reinsurance company; The customer exposure data and the regional meteorological data are input into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area. The catastrophic risk level includes the corresponding catastrophic loss amount. The claim-paying capacity data of each reinsurance company is input into the claim-paying capacity prediction model to obtain the claim-paying capacity level corresponding to each reinsurance company. The claim-paying capacity level includes the corresponding catastrophe payout amount. Based on the amount of catastrophe loss and the amount of catastrophe compensation, a target company is determined from the reinsurance companies, and the target company is used to enter into a reinsurance contract with the original insurance company.

2. The disaster recovery service processing method according to claim 1, characterized in that, Catastrophe risk levels include the corresponding reinsurance contribution ratio; After determining the target company from the reinsurance companies based on the aforementioned catastrophe risk level and the aforementioned payability level, the process further includes: Using blockchain technology, a smart contract is generated between the original insurance company and the target company based on the reinsurance sharing ratio; When a catastrophic loss occurs in the catastrophic area, the actual compensation amount for the catastrophic area is calculated so that the original insurance company and the target company can pay the actual compensation amount in accordance with the smart contract.

3. The disaster recovery service processing method according to claim 2, characterized in that, After calculating the catastrophic loss amount in the aforementioned catastrophic area so that the original insurance company and the target company can pay the catastrophic loss amount according to the smart contract, the process also includes: The amount of catastrophic loss, the amount of catastrophic loss, and the amount of actual compensation are input into the catastrophic compensation process audit model to obtain the compensation process audit result. The compensation process audit result is used to reflect the compliance of the compensation process and the rationality of the compensation result.

4. The disaster recovery service processing method according to claim 1, characterized in that, The customer exposure data and the regional meteorological data are input into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area, including: The regional meteorological data and the customer exposure data are preprocessed respectively to extract regional meteorological features from the regional meteorological data and regional customer features from the customer exposure data. The regional meteorological characteristics and the regional customer characteristics are fused to obtain the regional risk characteristics; The regional risk characteristics are input into the catastrophic risk prediction model to obtain the catastrophic risk level.

5. The disaster recovery service processing method according to claim 4, characterized in that, The regional meteorological data and the customer exposure data are preprocessed separately to extract regional meteorological features from the regional meteorological data and regional customer features from the customer exposure data, including: The regional meteorological data is cleaned, and the cleaned regional meteorological data is converted into regional meteorological features through feature engineering. The customer exposure data is cleaned, and the cleaned customer exposure data is transformed into regional customer characteristics through feature engineering.

6. The disaster recovery service processing method according to claim 4, characterized in that, The regional meteorological characteristics and the regional customer characteristics are fused to obtain regional risk characteristics, including: Based on the disaster-stricken area, spatial matching is performed on the meteorological characteristics of the area and the customer characteristics of the area to generate the risk characteristics of the area.

7. The disaster recovery service processing method according to any one of claims 1-6, characterized in that, The regional meteorological data includes: wind speed information, rainfall information, and temperature and humidity information; The customer exposure data includes: geographic information, asset information, risk resistance capabilities, and business continuity. The compensation capacity data includes: solvency, compensation records, and actual assets.

8. A disaster recovery service processing device, characterized in that, include: The data acquisition module is used to acquire regional meteorological data of the disaster area, customer exposure data of the original insurance company in the disaster area, and reimbursement capacity data of the reinsurance company. The risk prediction module is used to input the customer exposure data and the regional meteorological data into the catastrophic risk prediction model to obtain the catastrophic risk level of the catastrophic area, wherein the catastrophic risk level includes the corresponding catastrophic loss amount; The capacity prediction module is used to input the claim-paying capacity data of each reinsurance company into the claim-paying capacity prediction model to obtain the claim-paying capacity level of each reinsurance company, wherein the claim-paying capacity level includes the corresponding catastrophe claim amount. The company matching module is used to determine a target company from the reinsurance companies based on the amount of catastrophic loss and the amount of catastrophic compensation. The target company is used to sign a reinsurance contract with the original insurance company.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the catastrophe service processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the catastrophe service processing method according to any one of claims 1 to 7.

Images