A risk warning analysis method for property and cargo insurance claims based on precipitation disasters

Through careful analysis of precipitation disasters and the construction of risk index models, the problems of insufficient prospectiveness of precipitation disaster risk assessment and unconsidered risk differences in the existing technology are solved, and more accurate risk warning and more scientific risk are achieved manage.

CN118761846BActive Publication Date: 2024-11-22湖南省气象服务中心
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411238112.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-11-22
Estimated Expiration
2044-09-05

AI Technical Summary

Technical Problem

The existing technology has a forward-looking lack of analysis of wealth insurance claims in precipitation disasters, fails to fully consider the differences in insurance types, over-reliance on single threshold judgments, ignores geological factors and regional characteristics, resulting in inaccurate and passive response to risk warnings.

Method used

Divide the target area into sub-regions, collect and analyze precipitation data, evaluate the probability of precipitation disasters, combine urban drainage pipeline network and terrain data, build a precipitation factor risk index model for wealth insurance types, draw a risk level warning chart to provide more accurate and Targeted risk assessment.

Benefits of technology

It improves the scientificity and accuracy of precipitation disaster risk warning, enhances the disaster prevention and loss reduction capabilities of insurance institutions, and helps formulate more effective risk management strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761846B_ABST
    Figure CN118761846B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of disaster prevention and loss prevention of property insurance, and relates to a risk early warning analysis method for property and cargo insurance claims based on precipitation disasters. The method effectively evaluates the probability of occurrence of precipitation disasters in each target sub-region on the current rainy day. If the probability is greater than a set threshold, the numerical values ​​of each precipitation disaster factor corresponding to the current rainy day in each target sub-region are obtained, and the constructed claim risk index model based on precipitation factors for various property and cargo insurance types in each target sub-region is input. The claim risk index based on precipitation factors for various property and cargo insurance types in the current rainy day in each target sub-region is output, and after the risk level of claims for various property and cargo insurance types in the current rainy day in each target sub-region is obtained, a warning forecast map of the risk level of claims for property and cargo insurance in the target region is drawn for feedback, so as to provide more targeted insurance meteorological services for different property and cargo insurance types in different target sub-regions, and improve the disaster prevention and loss reduction capabilities of insurance institutions in responding to precipitation disaster risks in advance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of disaster prevention and loss prevention of property insurance, and relates to a risk early warning analysis method for property insurance claims based on precipitation disasters. Background Art

[0002] In the context of global climate change, extreme weather events occur frequently, especially precipitation disasters such as rainstorms, which not only threaten people's lives and safety, but also have a huge impact on the social economy, and cause serious property losses, thus posing a huge challenge to the insurance industry. Therefore, developing a scientific and effective early warning analysis method to identify the risk of property insurance claims caused by precipitation disasters is of great significance for insurance institutions to control risks, reduce losses, and improve market competitiveness.

[0003] In the prior art, there are also some solutions involving optimization and analysis of rainfall-related property insurance underwriting, for example, the invention patent application for typhoon and rainstorm risk rating method and related equipment based on artificial intelligence with Chinese application number CN2022109514058, which collects typhoon and rainstorm disaster data to obtain a typhoon and rainstorm data set to calculate the disaster factor risk index, and statistically analyzes the disaster factor risk index and insurance claims data of each typhoon and rainstorm disaster in history, and calculates the typhoon and rainstorm comprehensive risk index based on the disaster factor risk index and the insurance claims data, and the risk level of the typhoon and rainstorm is determined according to the typhoon and rainstorm comprehensive risk index and the insurance claims data. The typhoon and rainstorm areas to be risk rated are divided and underwriting and pricing are carried out accordingly. Although this has effectively improved the accuracy of underwriting and pricing for typhoon and rainstorm disasters, it still contains limitations that are common in existing technologies, namely: 1. Existing claims analysis technologies for property insurance applied to rainstorm disasters mostly focus on post-event response, that is, collecting data, assessing losses and making compensation after the disaster occurs. There are significant foresight deficiencies, which not only makes it difficult to effectively identify potential high-risk areas and time periods before a disaster occurs, and misses the opportunity to take preventive measures in advance, but also puts insurance institutions in a passive response state, increases claims costs, and reduces the efficiency and effectiveness of resource use.

[0004] 2. The existing claims analysis technology for property insurance applied to heavy rain and precipitation disasters fails to fully consider the differences in sensitivity of different types of insurance, such as engineering insurance, home insurance, and corporate insurance, to precipitation disasters. In actual operations, a one-size-fits-all risk assessment standard may be adopted, ignoring the specific risk characteristics of each type of property insurance and the diversity of insured objects, thereby affecting the accuracy and pertinence of the assessment of the impact of precipitation disasters on property insurance.

[0005] 3. The existing claims analysis technology for property insurance for rainstorm disasters relies too much on the "rainstorm" threshold defined by meteorological standards to determine whether it constitutes a claim risk, ignoring the complexity of the actual impact of precipitation events on property insurance users. On the one hand, precipitation below the "rainstorm" standard may also cause property losses due to factors such as local terrain and drainage systems. On the other hand, even if it reaches or exceeds the "rainstorm" standard, good protective measures and geographical location may protect some users from damage. This single threshold judgment method ignores the differences in geological factors, regional characteristics and urban construction protection levels, reducing the effectiveness of risk warning and the accuracy of claims processing. Summary of the invention

[0006] In view of this, in order to solve the problems raised in the above background technology, a risk warning analysis method for property and cargo insurance claims based on precipitation disasters is proposed.

[0007] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a risk warning analysis method for property insurance claims based on precipitation disasters, including: S1. Dividing the target area into sub-areas and marking them as target sub-areas, collecting precipitation data of the current rainy day in each target sub-area, including the precipitation amount, precipitation duration and precipitation intensity of each day on that day and within the future adjacent preset days.

[0008] S2. Extract the precipitation data corresponding to each historical precipitation disaster event in each target sub-area, including the precipitation amount, precipitation duration and precipitation intensity on the day of the event and the historical adjacent preset days, and evaluate the probability of precipitation disasters in each target sub-area on the current rainy day. If it is greater than the set threshold, the property and cargo insurance claim risk assessment instruction is triggered.

[0009] S3. Receive property and cargo insurance claims risk assessment instructions, obtain the values ​​of various precipitation disaster factors corresponding to the current rainy days in each target sub-area, input the constructed claims risk index model based on precipitation factors for various types of property and cargo insurance in each target sub-area, and output the claims risk index based on precipitation factors for various types of property and cargo insurance in each target sub-area on the current rainy days.

[0010] S4. Analyze the claims risk index range corresponding to each risk level of each type of property and cargo insurance in each target sub-region, and obtain the risk level of each type of property and cargo insurance in each target sub-region on the current rainy day.

[0011] S5. Based on the risk level of various types of property and cargo insurance claims on the current rainy days in each target sub-area, draw a warning forecast map of the risk level of property and cargo insurance claims in the target area and provide feedback.

[0012] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention compares precipitation data to understand the similarity between the current rainy day in each target sub-area and its historical precipitation disaster events, and effectively evaluates the precipitation disaster risk that may be caused by the current rainfall based on realistic evidence, thereby providing a more scientific and accurate basis for the claims of property and cargo insurance.

[0013] (2) The present invention calculates the precipitation disaster assessment correction factor of each target sub-region by combining the instantaneous pipe capacity of the urban construction drainage network of each target sub-region, the terrain data of each catchment area and the soil permeability data, and then reasonably corrects the probability of precipitation disaster occurrence on the current rainy day of each target sub-region, thereby ensuring the accuracy and reliability of the precipitation disaster occurrence probability analysis and providing strong support for subsequent claims risk warning and response.

[0014] (3) The present invention constructs a claim risk index model for each type of property and cargo insurance in each target sub-region based on precipitation factors through the relevant data on historical precipitation accident dates of each type of property and cargo insurance in each target sub-region, carefully considers the weight impact of each precipitation disaster factor on different types of property and cargo insurance, improves the defects of the prior art in analyzing the differences in the sensitivity of each type of property and cargo insurance to precipitation disasters, and improves the accuracy and pertinence of the claim risk assessment of property and cargo insurance affected by precipitation disasters.

[0015] (4) The present invention uses the optimal segmentation risk points corresponding to the risk levels of claims for each type of property and cargo insurance in each target sub-region as the entry point, analyzes the claim risk index range corresponding to the risk levels of claims for each type of property and cargo insurance in each target sub-region, and achieves more accurate claim risk assessment and more scientific claim risk management, which helps insurance institutions to more accurately price and formulate risk management strategies.

[0016] (5) The present invention provides feedback by drawing a warning forecast map of the risk level of property and cargo insurance claims in the target area, and more intuitively and accurately warns of the impact of precipitation risks on different property and cargo insurance types in different target sub-areas, thereby providing more targeted insurance meteorological services for different property and cargo insurance types in different target sub-areas, and improving the disaster prevention and loss reduction capabilities of insurance institutions in advance to deal with precipitation disaster risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 The present invention is a flowchart of the steps for implementing the method. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the present invention provides a risk warning analysis method for property insurance claims based on precipitation disasters, including: S1. Dividing the target area into sub-areas and marking them as target sub-areas, collecting precipitation data of the current rainy day in each target sub-area, including the precipitation amount, precipitation duration and precipitation intensity of the day and the adjacent preset days in the future.

[0021] S2. Extract the precipitation data corresponding to each historical precipitation disaster event in each target sub-area, including the precipitation amount, precipitation duration and precipitation intensity on the day of the event and the historical adjacent preset days, and evaluate the probability of precipitation disasters in each target sub-area on the current rainy day. If it is greater than the set threshold, the property and cargo insurance claim risk assessment instruction is triggered.

[0022] Specifically, the evaluation of the probability of precipitation disasters occurring in each target sub-area on the current rainy day includes: arranging in chronological order of date, with the current rainy day and each day within a preset number of days in the future constituting a current precipitation time series, and the day on which historical precipitation disaster events occurred and each day within a preset number of adjacent historical days constituting a historical precipitation time series of historical precipitation disaster events.

[0023] According to the principle of corresponding arrangement position in the current precipitation time series and the historical precipitation time series of historical precipitation disaster events, the current rainy day in each target sub-region and each day in the future preset days are matched with the day of each historical precipitation disaster event in the corresponding target sub-region and each day in the historical adjacent preset days. By comparing the precipitation data, the local precipitation similarity between the current rainy day in each target sub-region and each day in the future preset days and the corresponding day of each historical precipitation disaster event in the corresponding target sub-region is analyzed. The local precipitation similarity between the current rainfall in each target sub-region and its historical precipitation disaster events is obtained by mean calculation. ,in is the number of each target sub-region, , is the number of each historical precipitation disaster event, .

[0024] It should be noted that the analysis process of the above-mentioned local precipitation similarity is: calculate the absolute difference in precipitation between the current rainy day of a target sub-region and the corresponding day of each historical precipitation disaster event in the target sub-region, and take the ratio of the absolute difference to the precipitation on the current rainy day of the target sub-region as the deviation ratio of the precipitation on the current rainy day of the target sub-region. Similarly, obtain the deviation ratio of the precipitation duration and precipitation intensity on the current rainy day of the target sub-region, and take the inverse of the accumulated deviation ratio of precipitation, precipitation duration and precipitation intensity as the local precipitation similarity between the current rainy day of the target sub-region and the corresponding day of each historical precipitation disaster event in the target sub-region. It should be noted in particular that in order to prevent the accumulated value of the deviation ratio of precipitation, precipitation duration and precipitation intensity from being 0, a preset constant will be introduced into the accumulated value.

[0025] This is used to analyze the local precipitation similarity between the current rainy days in each target sub-region and the days within the preset number of days in the future and the days corresponding to the historical precipitation disaster events in the corresponding target sub-region.

[0026] The cumulative precipitation, cumulative precipitation duration and maximum precipitation intensity of the current precipitation time series of each target sub-region and its historical precipitation disaster events are counted, and the overall precipitation similarity between the current rainfall in each target sub-region and its historical precipitation disaster events is compared and analyzed. .

[0027] It should be noted that the analysis process of the above-mentioned overall precipitation similarity is: calculate the absolute difference between the current precipitation time series of each target sub-region and the historical precipitation time series of its various historical precipitation disaster events, the cumulative precipitation duration and the maximum precipitation intensity of the historical precipitation time series, and add it to the ratio of the cumulative precipitation, cumulative precipitation duration and maximum precipitation intensity of the current precipitation time series of the corresponding target sub-region and introduce the reciprocal after the preset constant is introduced as the overall precipitation similarity between the current rainfall in each target sub-region and its various historical precipitation disaster events.

[0028] By formula The probability of precipitation disasters occurring on the current rainy day in each target sub-area is obtained, where for , For the The precipitation disaster assessment correction factor for each target sub-area.

[0029] The embodiment of the present invention compares precipitation data to understand the degree of similarity between the current rainy day in each target sub-area and its historical precipitation disaster events, and effectively evaluates the precipitation disaster risk that may be caused by the current rainfall based on realistic evidence, thereby providing a more scientific and accurate basis for the claims of property and cargo insurance.

[0030] Specifically, the The acquisition process includes: collecting the instantaneous pipe capacity of the current urban construction drainage network in each target sub-area , combined with the cumulative precipitation of the current precipitation time series of each target sub-region and cumulative precipitation duration , analyze the drainage capacity coefficient of each target sub-area for the current rainy day.

[0031] It should be noted that the instantaneous pipeline capacity collection of the current urban construction drainage network in each of the above-mentioned target sub-areas is carried out by intelligent monitoring equipment equipped at the key nodes of each pipeline in the urban construction drainage network, such as a flow meter, to obtain the volume of fluid flowing through the pipeline cross-section per unit time at each key node of the current pipeline in each target sub-area, and the volume of fluid flowing through the pipeline cross-section per unit time in the current urban construction drainage network in each target sub-area is obtained by mean calculation, which is used as the instantaneous pipeline capacity of the current urban construction drainage network in each target sub-area.

[0032] According to the watershed divisions marked by the urban construction drainage network in each target sub-area, the slopes of the current watershed divisions in each target sub-area are collected. , maximum depression depth and soil water permeability , is the number of each catchment area, , analyze the water storage capacity coefficient of each target sub-area for the current rainy day.

[0033] It should be noted that the watershed areas marked by the urban construction drainage network of each target sub-area mentioned above are obtained by referring to the urban construction drainage network map of each target sub-area. The slope, maximum depression depth and soil permeability of each current watershed area of ​​each target sub-area are extracted from the latest geological survey report data of the urban construction drainage network website of each target sub-area.

[0034] The ratio of the drainage capacity coefficient to the water storage capacity coefficient is used as the precipitation response capacity coefficient of each target sub-area for the current rainy day. , and the reasonable threshold of regional precipitation response capacity coefficient stored in the WEB cloud Comparison is performed by computational model Obtain the precipitation disaster assessment correction factor for each target sub-area, is a natural constant.

[0035] Specifically, the analysis formula of the drainage capacity coefficient of each target sub-area for the current rainy day is: ,in For WEB cloud storage The preset area of ​​each target sub-area.

[0036] Specifically, the analysis formula of the water storage capacity coefficient of each target sub-area for the current rainy day is: , They are the preset reference slope, reference depression depth and reference soil permeability respectively.

[0037] The embodiment of the present invention calculates the precipitation disaster assessment correction factor of each target sub-area by combining the instantaneous pipe capacity of the urban construction drainage network of each target sub-area, the terrain data of each catchment area and the soil permeability data, and then reasonably corrects the probability of precipitation disaster occurrence on the current rainy day of each target sub-area, thereby ensuring the accuracy and reliability of the precipitation disaster occurrence probability analysis, and providing strong support for subsequent claims risk warning and response.

[0038] S3. Receive property and cargo insurance claims risk assessment instructions, obtain the values ​​of various precipitation disaster factors corresponding to the current rainy days in each target sub-area, input the constructed claims risk index model based on precipitation factors for various types of property and cargo insurance in each target sub-area, and output the claims risk index based on precipitation factors for various types of property and cargo insurance in each target sub-area on the current rainy days.

[0039] Specifically, the obtaining of the numerical values ​​of each precipitation disaster factor corresponding to the current rainy day of each target sub-area includes: obtaining the 24-hour cumulative precipitation, 48-hour cumulative precipitation, 72-hour cumulative precipitation and the previous day's maximum rainfall intensity corresponding to the current rainy day of each target sub-area through the forecast content of the meteorological station, as the numerical values ​​of each precipitation disaster factor corresponding to the current rainy day of each target sub-area, wherein the starting time point of the 24-hour cumulative precipitation is the preset time point of the current rainy day, and the starting time points of the 48-hour cumulative precipitation and the 72-hour cumulative precipitation are the preset time points of the day before the current rainy day.

[0040] Exemplarily, if the current rainy day is November 13, the corresponding precipitation disaster factors include: the starting time for reporting the 24-hour cumulative precipitation is 8:00 on November 13, which is the cumulative precipitation from 08:00 on November 13 to 08:00 on November 14; the starting time for reporting the 48-hour cumulative precipitation and the 72-hour cumulative precipitation is 8:00 on November 12, which are the cumulative precipitation from 08:00 on November 12 to 08:00 on November 14, and the cumulative precipitation from 08:00 on November 12 to 08:00 on November 15, respectively; the maximum rainfall intensity of the previous day is the maximum rainfall per unit time from 08:00 on November 12 to 08:00 on November 13.

[0041] Specifically, the process of constructing the claims risk index model for each type of property and cargo insurance in each target sub-region based on precipitation factors includes: randomly selecting a type of property and cargo insurance in a target sub-region, and establishing an initial model for the type of property and cargo insurance in the target sub-region, which includes a preset number of prediction factors that affect the number of claims for this type of property and cargo insurance, but does not include various precipitation disaster factors.

[0042] The number of risk points, the values ​​of each precipitation disaster factor and the values ​​of each prediction factor corresponding to each historical precipitation risk date of this type of property and cargo insurance in the target sub-area are extracted from the WEB cloud, and the values ​​of each prediction factor corresponding to each historical precipitation risk date are substituted into the initial model. The initial model outputs the initial predicted number of risk points for each historical precipitation risk date of this type of property and cargo insurance in the target sub-area.

[0043] It should be noted that the above-mentioned prediction factors refer to data variables that are related to or independent of precipitation disaster factors and have an impact on the claims risk of property and cargo insurance, such as wind speed, humidity, temperature, and topography. These prediction factors are set in advance by the system before modeling and serve as an important part of the model. Their role is to further improve and refine the construction of the claims risk index model based on precipitation factors for various types of property and cargo insurance in each target sub-region. The introduction and consideration of these prediction factors aims to improve the rationality and accuracy of the model, so as to more accurately evaluate and predict the impact of precipitation disasters on the claims risk of property and cargo insurance.

[0044] The numerical value of a certain precipitation hazard factor corresponding to each historical precipitation accident date of this type of property and cargo insurance in the target sub-region is added to the initial model, and the initial model outputs the first-order predicted number of accident points for each historical precipitation accident date of this type of property and cargo insurance in the target sub-region after the precipitation hazard factor is added. The average value-added contribution of the precipitation hazard factor to the prediction of the accident point after being added to the initial model is compared and analyzed, and the next precipitation hazard factor is added on the basis of the precipitation hazard factor added to the initial model. Similarly, the average value-added contribution of the next precipitation hazard factor to the prediction of the accident point after being added to the initial model is analyzed, thereby obtaining the average value-added contribution of each precipitation hazard factor to the prediction of the accident point of this type of property and cargo insurance in the target sub-region, and the percentage of the total value-added contribution of precipitation hazard factors to the prediction of the accident point is taken as the contribution percentage of each precipitation hazard factor of this type of property and cargo insurance in the target sub-region.

[0045] It should be noted that the specific process of the above-mentioned comparative analysis of the average value-added contribution to the risk point prediction after the precipitation disaster factor is added to the initial model is as follows: the first-order predicted risk points of each historical precipitation risk date for this type of property and cargo insurance in the target sub-area after the precipitation disaster factor is added output by the initial model are subtracted from the initial predicted risk points of each historical precipitation risk date for this type of property and cargo insurance in the target sub-area output by the initial model, so as to obtain the enhanced predicted risk points of each historical precipitation risk date for this type of property and cargo insurance in the target sub-area, and the average value-added contribution to the risk point prediction after the precipitation disaster factor is added to the initial model is obtained by mean calculation.

[0046] Similarly, the contribution percentage of each precipitation disaster factor for each type of property insurance in each target sub-area is obtained. , The serial numbers of various types of property and cargo insurance. , is the number of each precipitation disaster factor, .

[0047] The precipitation hazard factors of each target sub-area on rainy days are marked as , thus constructing the claim risk index model based on precipitation factors for various types of property insurance in each target sub-region ,in The number of types of property and cargo insurance.

[0048] The embodiment of the present invention constructs a claim risk index model based on precipitation factors for various types of property and cargo insurance in each target sub-region through relevant data on historical precipitation accident dates for various types of property and cargo insurance in each target sub-region, carefully considers the weight influence of various precipitation disaster factors on different types of property and cargo insurance, improves the defects of the prior art in analyzing the differences in the sensitivity of various types of property and cargo insurance to precipitation disasters, and improves the accuracy and pertinence of claim risk assessment for property and cargo insurance affected by precipitation disasters.

[0049] S4. Analyze the claims risk index range corresponding to each risk level of each type of property and cargo insurance in each target sub-region, and obtain the risk level of each type of property and cargo insurance in each target sub-region on the current rainy day.

[0050] Specifically, the analysis of the corresponding claims risk index intervals of each risk level of each type of property and cargo insurance in each target sub-region includes: extracting the values ​​of precipitation disaster factors corresponding to the best segmentation risk points of each risk level of each type of property and cargo insurance in each target sub-region based on the best segmentation risk points stored in the WEB cloud, inputting the values ​​into the claims risk index model based on precipitation factors of each type of property and cargo insurance in each target sub-region, outputting the claims risk index of each historical precipitation risk date corresponding to the best segmentation risk points of each risk level of each type of property and cargo insurance in each target sub-region, and obtaining the best segmentation claim risk index corresponding to each risk level of each type of property and cargo insurance in each target sub-region by averaging the claims risk indexes of each historical precipitation risk date, thereby obtaining the claims risk index intervals corresponding to each risk level of each type of property and cargo insurance in each target sub-region.

[0051] Specifically, the method of obtaining the risk level of claims for various types of property and cargo insurance on the current rainy day in each target sub-region includes: outputting the claim risk level corresponding to the claim risk index interval of the corresponding property and cargo insurance type in the corresponding target sub-region based on the claim risk index of various types of property and cargo insurance on the current rainy day in each target sub-region as the risk level of claims for various types of property and cargo insurance on the current rainy day in each target sub-region.

[0052] The embodiment of the present invention takes the optimal segmentation risk points corresponding to each risk level of each type of property and cargo insurance in each target sub-region as the entry point, analyzes the claim risk index range corresponding to each risk level of each type of property and cargo insurance in each target sub-region, achieves more accurate claim risk assessment and more scientific claim risk management, and helps insurance institutions to more accurately price and formulate risk management strategies.

[0053] S5. Based on the risk level of various types of property and cargo insurance claims on the current rainy days in each target sub-area, draw a warning forecast map of the risk level of property and cargo insurance claims in the target area and provide feedback.

[0054] Specifically, the method of drawing a warning forecast map of the risk level of property and cargo insurance claims in the target area includes: marking various types of property and cargo insurance with different shapes, marking each risk level of property and cargo insurance with different colors, and marking the corresponding risk level of property and cargo insurance for each target sub-area on the current rainy day in the map of the target area to form a warning forecast map.

[0055] The embodiment of the present invention provides feedback by drawing a warning forecast map of the risk level of property and cargo insurance claims in the target area, and more intuitively and accurately warns that different property and cargo insurance types in different target sub-areas are affected by precipitation risks, thereby providing more targeted insurance meteorological services for different property and cargo insurance types in different target sub-areas, and improving the disaster prevention and loss reduction capabilities of insurance institutions to respond to precipitation disaster risks in advance.

[0056] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. Risk warning analysis method for property insurance claims based on precipitation disasters, It is characterized in that The method includes: S1. Divide the target area into sub-areas and mark them as target sub-areas, collect precipitation data of the current rainy day in each target sub-area, including the precipitation amount, precipitation duration and precipitation intensity of each day on the current day and within the next preset number of adjacent days; S2. Extract the precipitation data corresponding to each historical precipitation disaster event in each target sub-area, including the precipitation amount, precipitation duration and precipitation intensity on the day of the event and the preset number of adjacent historical days, and evaluate the probability of precipitation disasters in each target sub-area on the current rainy day. If it is greater than the set threshold, the property and cargo insurance claim risk assessment instruction is triggered; S3. Receive a property insurance claim risk assessment instruction, obtain the values ​​of each precipitation hazard factor corresponding to the current rainy day in each target sub-region, input the constructed claim risk index model based on precipitation hazard factors for each type of property insurance in each target sub-region, and output the claim risk index based on precipitation hazard factors for each type of property insurance in each target sub-region on the current rainy day; S4. Analyze the claims risk index intervals corresponding to the risk levels of various types of property and cargo insurance in each target sub-region, and obtain the risk levels of various types of property and cargo insurance in each target sub-region on the current rainy day; S5. Based on the risk level of claims for various types of property and cargo insurance in each target sub-region on the current rainy day, draw a warning forecast map of the risk level of property and cargo insurance claims for the target region and provide feedback; The obtaining of the numerical values ​​of each precipitation disaster factor corresponding to the current rainy day of each target sub-region includes: obtaining the 24-hour cumulative precipitation, 48-hour cumulative precipitation, 72-hour cumulative precipitation and the maximum rainfall intensity of the previous day corresponding to the current rainy day of each target sub-region through the forecast content of the meteorological station, as the numerical values ​​of each precipitation disaster factor corresponding to the current rainy day of each target sub-region, wherein the starting time point of the 24-hour cumulative precipitation is the preset time point of the current rainy day, and the starting time points of the 48-hour cumulative precipitation and the 72-hour cumulative precipitation are the preset time points of the day before the current rainy day; The construction process of the claim risk index model of each type of property insurance in each target sub-region based on precipitation hazard factors includes: randomly selecting a type of property insurance in a target sub-region, and establishing an initial model for the type of property insurance in the target sub-region, which includes a preset number of prediction factors that affect the number of claims of the type of property insurance, but does not include the precipitation hazard factors; Extract the number of risk points, the values ​​of each precipitation hazard factor and the values ​​of each prediction factor corresponding to each historical precipitation risk date of this type of property and cargo insurance in the target sub-region from the WEB cloud, substitute the values ​​of each prediction factor corresponding to each historical precipitation risk date into the initial model, and the initial model outputs the initial predicted number of risk points for each historical precipitation risk date of this type of property and cargo insurance in the target sub-region; Add the value of a certain precipitation hazard factor corresponding to each historical precipitation accident date of this type of property and cargo insurance in the target sub-region to the initial model, and the initial model outputs the first-order predicted number of accident points for each historical precipitation accident date of this type of property and cargo insurance in the target sub-region after the precipitation hazard factor is added, compare and analyze the average value-added contribution of the precipitation hazard factor to the prediction of the accident point after it is added to the initial model, and add the next precipitation hazard factor on the basis of the precipitation hazard factor added to the initial model, and similarly analyze the average value-added contribution of the next precipitation hazard factor to the prediction of the accident point after it is added to the initial model, thereby obtaining the average value-added contribution of each precipitation hazard factor to the prediction of the accident point of this type of property and cargo insurance in the target sub-region, and use the percentage of the total value-added contribution of precipitation hazard factors to the prediction of the accident point as the contribution percentage of each precipitation hazard factor of this type of property and cargo insurance in the target sub-region; Similarly, the contribution percentage of each precipitation disaster factor for each type of property insurance in each target sub-area is obtained. , The serial numbers of various types of property and cargo insurance. , is the number of each precipitation disaster factor, ; The precipitation hazard factors of each target sub-area on rainy days are marked as , thus constructing the claim risk index model for various types of property insurance in each target sub-region based on precipitation disaster factors ,in is the number of types of property and cargo insurance; The analysis of the corresponding claims risk index intervals of each risk level of each type of property and cargo insurance in each target sub-region includes: extracting the values ​​of each precipitation hazard factor corresponding to the best segmentation risk points of each risk level of each type of property and cargo insurance in each target sub-region stored in the WEB cloud, inputting the values ​​into the claims risk index model based on precipitation hazard factors of each type of property and cargo insurance in each target sub-region, outputting the claims risk index corresponding to the best segmentation risk points of each risk level of each type of property and cargo insurance in each target sub-region and corresponding to each historical precipitation risk date, and obtaining the best segmentation claim risk index corresponding to each risk level of each type of property and cargo insurance in each target sub-region by averaging the claims risk index of each historical precipitation risk date, thereby obtaining the claims risk index intervals corresponding to each risk level of each type of property and cargo insurance in each target sub-region; The method of obtaining the risk level of claims for various types of property and cargo insurance on the current rainy day in each target sub-region includes: outputting the claim risk level corresponding to the claim risk index interval of the corresponding property and cargo insurance type in the corresponding target sub-region based on the claim risk index of various types of property and cargo insurance on the current rainy day in each target sub-region as the risk level of claims for various types of property and cargo insurance on the current rainy day in each target sub-region.

2. The method for early warning and analysis of property and cargo insurance claims based on precipitation disasters according to claim 1, Features: The evaluation of the probability of occurrence of precipitation disasters in each target sub-area on the current rainy day includes: arranging in chronological order of date, the current rainy day and each day within a preset number of days in the future constitute a current precipitation time series, and the day on which a historical precipitation disaster event occurred and each day within a preset number of adjacent historical days constitute a historical precipitation time series of historical precipitation disaster events; According to the principle of corresponding arrangement position in the current precipitation time series and the historical precipitation time series of historical precipitation disaster events, the current rainy day in each target sub-region and each day within the preset number of future days are matched with the day of each historical precipitation disaster event in the corresponding target sub-region and each day within the preset number of historical days. By comparing the precipitation data, the local precipitation similarity between the current rainy day in each target sub-region and each day within the preset number of future days and the corresponding day of each historical precipitation disaster event in the corresponding target sub-region is analyzed. The local precipitation similarity between the current rainfall in each target sub-region and its historical precipitation disaster events is obtained by average calculation. ,in is the number of each target sub-region, , is the number of each historical precipitation disaster event, ; The cumulative precipitation, cumulative precipitation duration and maximum precipitation intensity of the current precipitation time series of each target sub-region and its historical precipitation disaster events are counted, and the overall precipitation similarity between the current rainfall in each target sub-region and its historical precipitation disaster events is compared and analyzed. ; By formula The probability of precipitation disasters occurring on the current rainy day in each target sub-area is obtained, where for , For the The precipitation disaster assessment correction factor for each target sub-area.

3. The method for early warning and analysis of property and cargo insurance claims based on precipitation disasters according to claim 2, Features: Said The acquisition process includes: collecting the instantaneous pipe capacity of the current urban construction drainage network in each target sub-area , combined with the cumulative precipitation of the current precipitation time series of each target sub-region and cumulative precipitation duration , analyze the drainage capacity coefficient of each target sub-area for the current rainy day; According to the watershed divisions marked by the urban construction drainage network in each target sub-area, the slopes of the current watershed divisions in each target sub-area are collected. , maximum depression depth and soil water permeability , is the number of each catchment area, , analyze the water storage capacity coefficient of each target sub-area for the current rainy day; The ratio of the drainage capacity coefficient to the water storage capacity coefficient is used as the precipitation response capacity coefficient of each target sub-area for the current rainy day. , and the reasonable threshold of regional precipitation response capacity coefficient stored in the WEB cloud Comparison is performed by computational model Obtain the precipitation disaster assessment correction factor for each target sub-area, is a natural constant.

4. The method for early warning and analysis of property and cargo insurance claims based on precipitation disasters according to claim 3, Features: The analysis formula of the drainage capacity coefficient of each target sub-area for the current rainy day is: ,in For WEB cloud storage The preset area of ​​each target sub-area.

5. The method for early warning and analysis of property and cargo insurance claims based on precipitation disasters according to claim 3, Features: The analysis formula of the water storage capacity coefficient of each target sub-area for the current rainy day is: , They are the preset reference slope, reference depression depth and reference soil permeability respectively.

6. The method for early warning and analysis of property and cargo insurance claims based on precipitation disasters according to claim 1, Features: The method of drawing a warning forecast map of the risk level of property and cargo insurance claims in the target area includes: marking various types of property and cargo insurance with different shapes, marking each risk level of property and cargo insurance claims with different colors, and marking the corresponding risk level of property and cargo insurance claims for each target sub-area on the current rainy day in the map of the target area to form a warning forecast map.