Agricultural risk operation management method based on crop knowledge graph

Abstract

The present invention discloses a method for controlling agricultural insurance operations based on a crop knowledge graph, comprising the following steps: S1) constructing an agricultural crop knowledge graph based on agricultural insurance crop data, meteorological disaster data, and pest and disease data; S2) integrating the knowledge graph data into the underwriting and claims process; S3) in the underwriting link, first verify whether the corresponding crop is planted in the area where the insurance policy is located, whether the crop is insured before the emergence period, and use the current agricultural insurance points of the area to assist in scoring and grading the policy, thereby determining the underwriting risk level; S4) in the claims link, through on-site investigation of the meteorological disaster data and / or pest and disease data of the area, assist in verifying whether the claim is caused by the corresponding disaster; and according to the growth period data, combined with the reporting time and the terms and conditions, finally determine whether the compensation ratio is reasonable. The agricultural insurance operation control method based on the crop knowledge graph provided by the present invention can carry out refined control of underwriting and claims in a multi-dimensional and scientific manner.

Description

Technical Field

[0001] The present invention relates to a method for controlling agricultural insurance operations, and in particular to a method for controlling agricultural insurance operations based on a crop knowledge graph. Background Art

[0002] The existing crop growth period, reproductive period, and real-time data are all independent data and are not integrated with the underwriting and claims settlement business of agricultural insurance.

[0003] In order to refine the underwriting and claims settlement management and control of crop insurance during underwriting and payment, and implement the central government's spirit of accurate underwriting and accurate claims settlement, it is necessary to introduce crop knowledge graphs in the underwriting and claims settlement process, strengthen underwriting management and control, and implement claims risk reduction. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an agricultural insurance operation management and control method based on crop knowledge graph, which can carry out refined management and control of underwriting and claims in a multi-dimensional and scientific manner.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems is to provide an agricultural insurance operation management method based on a crop knowledge graph, comprising the following steps: S1) constructing an agricultural crop knowledge graph based on agricultural insurance crop data, meteorological disaster data and pest and disease data; S2) integrating the knowledge graph data into the underwriting and claims process; S3) in the underwriting stage, first use the agricultural crop knowledge graph to check whether the corresponding crops are planted in the area where the insurance policy is located, whether the crop is insured before the seedling stage, and use the current agricultural insurance points in the area to assist in scoring and grading the policy, so as to determine the underwriting risk level; S4) in the claims stage, through on-site investigation of the meteorological disaster data and / or pest and disease data in the area, assist in verifying whether the claim is caused by the corresponding disaster; and based on the growth period data, combined with the reporting time and the terms and conditions, finally determine whether the compensation ratio is reasonable.

[0006] Furthermore, step S1 forms a crop growth period data map and a yield area data map for provinces, cities and counties across the country by combining the yearbook data published annually by agricultural meteorological stations and provinces, cities and counties with data from the Ministry of Agriculture and Rural Affairs, the Chinese Academy of Agricultural Sciences and some areas of the Agricultural Bureau.

[0007] Furthermore, step S3 combines historical meteorological data, crops, growth periods and disaster factors, calculates various disaster indexes based on meteorology, and then converts the various indexes into the final agricultural insurance score for the crop through the hierarchical analysis method. When the score is less than the preset threshold, it means that there is a greater planting risk in the area for planting the crop, and an early warning prompt is added during the underwriting process.

[0008] Furthermore, the step S3 forms a risk score through various disaster algorithm indicators; derives the meteorological yield and trend yield through historical yield area data combined with meteorological factors, and forms a vulnerability score; forms an exposure score through historical cultivated land area, and finally forms the final agricultural insurance score through the hierarchical analysis method.

[0009] Furthermore, the disaster algorithm indicators in step S3 include a water deficit deviation index, a dry hot wind level index and a waterlogging index.

[0010] Furthermore, the step S3 sets respective weights for the risk score, vulnerability score and exposure score; the final agricultural insurance score is: 100-(((1-risk / 100)×risk weight)×((1-exposure / 100)×exposure weight)×((1-vulnerability / 100)×vulnerability weight))×100.

[0011] Furthermore, step S3 also includes combining academic research with actual historical meteorological and pest data to find out the meteorological factors triggered by pests and diseases in different growth periods; starting from the growth period when the pests and diseases will occur, combined with the historical meteorological data of 15 days, it is determined whether the meteorological factors appear continuously. If so, an influence coefficient is formed according to the number of consecutive days of appearance; subsequent daily rolling calculations are performed, and the influence coefficients of multiple meteorological factors are combined to form the final pest and disease prevention level.

[0012] Furthermore, the prevention level of the corn crop blight disease in step S3 is graded as follows:

[0013] Level 1: Meteorological conditions are very suitable for the occurrence of corn leaf blight; temperature is 20℃~26℃, relative humidity is ≥90%, or there is dew on the leaves, precipitation in June is ≥80 mm, precipitation in July is ≥80 mm, precipitation in August is 100~150 mm, if any of the above conditions is not met, the level will be reduced by one level;

[0014] Level 2: Meteorological conditions are relatively suitable for the occurrence of corn leaf blight, with temperatures ranging from 5 to 20°C or 26 to 35°C, precipitation in June less than 80 mm, and precipitation in July less than 80 mm;

[0015] Level 3: Meteorological conditions are not suitable for the occurrence of corn leaf blight, temperature <5℃ or temperature >35℃.

[0016] Compared with the prior art, the present invention has the following beneficial effects: the agricultural insurance operation management method based on the crop knowledge graph provided by the present invention, through the construction of the knowledge graph of agricultural insurance crops, from basic data such as crop type, growth period, yield area, to agricultural insurance points, agricultural meteorological disasters, pest and disease forecasts and other measurement data formed according to real-time meteorological data combined with the measurement model; thereby, it is possible to carry out refined management and control of underwriting and claims in a multi-dimensional and scientific manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of agricultural insurance operation management and control based on crop knowledge graph of the present invention;

[0018] Figure 2 A schematic diagram of the knowledge graph structure provided by the present invention;

[0019] Figure 3 This is a schematic diagram of calculating a single disaster index according to the present invention. DETAILED DESCRIPTION

[0020] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0021] Figure 1 This is a flow chart of agricultural insurance operation management and control based on the crop knowledge graph of the present invention.

[0022] See also Figure 1 The agricultural insurance operation management method based on the crop knowledge graph provided by the present invention comprises the following steps:

[0023] S1) Constructing agricultural crop knowledge graph based on agricultural crop data, meteorological disaster data and pest and disease data;

[0024] S2) Integrate knowledge graph data into the underwriting and claims process;

[0025] S3) In the underwriting stage, the agricultural crop knowledge map is used to check whether the corresponding crops are planted in the area where the insurance policy is located, whether the crops are insured before the seedling stage, and the current agricultural insurance score of the area is used to assist in scoring and grading the policy, thereby determining the underwriting risk level;

[0026] S4) In the claims process, the meteorological disaster data and / or pest and disease data of the area are surveyed on site to assist in verifying whether the claims are caused by the corresponding disasters; and based on the growth period data, combined with the reporting time and the terms and conditions, it is finally determined whether the compensation ratio is reasonable.

[0027] The agricultural crop knowledge graph used in this invention is a comprehensive service based on agricultural crop, weather, disaster and other data. It aims to provide a basis and early warning for underwriting and claims through various agricultural weather data combined with various algorithm models, such as Figure 2 shown.

[0028] The present invention is built on existing meteorological data and some algorithms, and will be expanded in the future according to business needs, adding more data and advanced algorithm services, and forming a set of agricultural crop knowledge services based on business and data.

[0029] ① Through the yearbook data published by agricultural meteorological stations and provinces, cities and counties each year, combined with data from the Ministry of Agriculture and Rural Affairs, the Chinese Academy of Agricultural Sciences, and some areas of the Agricultural Bureau, a national crop growth period data map and a relatively accurate yield area data map are formed for provinces, cities, and counties across the country.

[0030] ② Through academic papers from various regions, combined with meteorological disaster data and pest and disease data, an algorithm for agricultural meteorological disasters and pest and disease suitable for various crops is formed.

[0031] The following is an algorithm for the occurrence of corn leaf blight based on the following academic research combined with weather forecasts and historical data.

[0032] Level 1: The weather conditions are very suitable for the occurrence of corn leaf blight. The temperature is 20℃~26℃, the relative humidity is ≥90%, or there is dew on the leaves, the precipitation in June is ≥80 mm, the precipitation in July is ≥80 mm, and the precipitation in August is 100~150 mm. If any of the above conditions is not met, it will be downgraded to one level.

[0033] Level 2: Meteorological conditions are relatively suitable for the occurrence of corn leaf blight. Temperature is 5-20℃ or 26-35℃, precipitation in June is less than 80 mm, precipitation in July is less than 80 mm

[0034] Level 3: Weather conditions are not suitable for the occurrence of corn leaf blight. Temperature < 5℃ or temperature > 35℃.

[0035] ③ Combine historical meteorological data, crops, growth period, disasters and other factors, calculate various disaster indexes according to the weather, and then convert each index into the final agricultural insurance score (0-100 points) for crops through the hierarchical analysis method. When the score is less than 60, it means that the area will have a greater planting risk for the crop, so as to increase early warning reminders in the underwriting process, warn the underwriters, and control the order process with more control; and subsequently implement mandatory wind survey activities during insurance monitoring. It can be used as an auxiliary basis in underwriting and claims. And the disaster index in the calculation process is extracted separately to form the hazard score of a single crop in a certain area to a certain meteorological disaster.

[0036] Through the above three types of data (yield area data + daily meteorological and agricultural disaster data; crop pest and disease data; agricultural insurance data for regions and crops), the knowledge graph data is integrated into the underwriting and claims process. A set of knowledge graph rule mechanisms based on agricultural insurance is formed.

[0037] In the underwriting stage, before underwriting, the crop knowledge graph can be used to determine whether the crop is planted in the area where the insurance policy is located, whether the crop is insured before the seedling stage, and the current agricultural insurance score of the area can be used to assist in scoring and grading the policy, thereby determining the underwriting risk level.

[0038] In the claim settlement process, we conduct on-site inspections and tests on agricultural weather disasters, pests and diseases in the area to verify whether the corresponding disasters have caused claims. And based on the growth period data, combined with the reporting time and the terms and conditions, we can finally determine whether the compensation ratio is reasonable.

[0039] The agricultural insurance operation management method based on crop knowledge graph provided by the present invention has the following substantial characteristics and beneficial effects:

[0040] 1. A set of basic meteorological indicator rules for different crops in different growth periods have been formed, and the results are calculated based on the rules.

[0041] 2. Based on the growth period data of each region, it can help the claims adjuster determine the growth period of the crops at the time of filing the report, and the auxiliary system can calibrate and verify the compensation ratio.

[0042] 3. Combine academic research with actual historical meteorological and pest data to find out the meteorological factors that are triggered by pests and diseases in different growth periods. Starting from the growth period when pests and diseases will occur, combined with the historical meteorological data of 15 days, determine whether the meteorological factors appear continuously. If so, form an impact coefficient based on the number of consecutive days of appearance. Subsequent daily rolling calculations are performed, and the impact coefficients of multiple meteorological factors are combined to form the final pest and disease prevention level to achieve an early warning effect. Provide value-added services to farmers on the C-end, combine early warnings and farming advice, remind farmers of planting risks, and take early prevention measures. On the E-end, a map-based early warning function is also formed according to business needs, allowing branch companies to promptly discover risks within their coverage.

[0043] 4. Combined with the historical meteorological data of various places over the past 30 years, various disaster algorithm indicators (such as water deficit anomaly index, dry hot wind grade index, waterlogging index, etc.) are used to form a risk score; historical yield area data, combined with meteorological and other factors, the meteorological yield and trend yield are obtained to finally form a vulnerability score; the exposure score is formed through the historical cultivated land area, and the above three scores are formed into the final agricultural risk score through the hierarchical analysis method, such as Figure 3 shown.

[0044] On the underwriting side, before underwriting, the risk level is confirmed based on the agricultural insurance points of the crop in the area within the insured time range to determine whether it is necessary to strengthen the approval level, increase the verification materials, or even confirm whether to refuse insurance.

[0045] Due to differences in weather and environment in different regions, the scores obtained by the same algorithm will have a large difference in relationship with the compensation data. Therefore, in implementation, the present invention can also continuously make further adjustments to the algorithm based on the compensation data, form an independent disaster definition for each region, and the proportion of each disaster score in the final agricultural insurance score and risk score.

[0046] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the definition of the claims.

Claims

1. A method for agricultural insurance operation management and control based on crop knowledge graph, characterized in that: The steps include: S1) Constructing agricultural crop knowledge graph based on agricultural crop data, meteorological disaster data and pest and disease data; S2) Integrate knowledge graph data into the underwriting and claims process; S3) In the underwriting stage, the agricultural crop knowledge map is used to check whether the corresponding crops are planted in the area where the insurance policy is located, whether the crops are insured before the seedling stage, and the current agricultural insurance score of the area is used to assist in scoring and grading the policy, thereby determining the underwriting risk level; S4) In the claim settlement process, by conducting on-site surveys of the meteorological disaster data and / or pest and disease data in the area, it is helpful to verify whether the claim has been made due to the corresponding disaster; And based on the reproductive period data, combined with the reporting time and clause settings, it is ultimately determined whether the compensation ratio is reasonable.

2. The agricultural insurance operation management method based on crop knowledge graph according to claim 1, characterized in that: The step S1 forms a crop growth period data map and a yield area data map for provinces, cities and counties across the country by combining the yearbook data published annually by agricultural meteorological stations and provinces, cities and counties with data from the Ministry of Agriculture and Rural Affairs, the Chinese Academy of Agricultural Sciences and some areas of the Agricultural Bureau.

3. The agricultural insurance operation management method based on crop knowledge graph according to claim 1, characterized in that: The step S3 combines historical meteorological data, crops, growth periods and disaster factors, calculates various disaster indexes according to the weather, and then converts the various indexes into the final agricultural insurance score for the crop through the hierarchical analysis method. When the score is less than the preset threshold, it means that there is a greater planting risk in the area for planting the crop, and an early warning prompt is added during the underwriting process.

4. The agricultural insurance operation management method based on crop knowledge graph according to claim 3 is characterized in that: The step S3 forms a risk score through various disaster algorithm indicators; combines historical yield area data with meteorological factors to obtain meteorological yield and trend yield, and forms a vulnerability score; forms an exposure score through historical cultivated land area, and finally forms the final agricultural risk score through the hierarchical analysis method.

5. The agricultural insurance operation management method based on crop knowledge graph according to claim 4 is characterized in that: The disaster algorithm indicators in step S3 include a water deficit deviation index, a dry hot wind level index and a waterlogging index.

6. The agricultural insurance operation management method based on crop knowledge graph according to claim 4, characterized in that: The step S3 sets respective weights for the risk score, vulnerability score and exposure score; the final agricultural insurance score is: 100-(((1-risk / 100)×risk weight)×((1-exposure / 100)×exposure weight)×((1-vulnerability / 100)×vulnerability weight))×100.

7. The agricultural insurance operation management method based on crop knowledge graph according to claim 1, characterized in that: The step S3 also includes combining academic research with actual historical meteorological and pest data to find out the meteorological factors triggered by pests and diseases in different growth periods; starting from the growth period when the pests and diseases will occur, combined with the historical meteorological data of 15 days, it is determined whether the meteorological factors appear continuously. If so, an influence coefficient is formed according to the number of consecutive days of appearance; subsequent daily rolling calculations are performed, and the influence coefficients of multiple meteorological factors are combined to form the final pest and disease prevention level.

8. The agricultural insurance operation management method based on crop knowledge graph according to claim 7, characterized in that: The prevention level of the corn crop blight disease in step S3 is graded as follows: Level 1: Meteorological conditions are very suitable for the occurrence of corn leaf blight; temperature is 20℃~26℃, relative humidity is ≥90%, or there is dew on the leaves, precipitation in June is ≥80 mm, precipitation in July is ≥80 mm, precipitation in August is 100~150 mm, if any of the above conditions is not met, the level will be reduced by one level; Level 2: Meteorological conditions are relatively suitable for the occurrence of corn leaf blight, with temperatures ranging from 5 to 20°C or 26 to 35°C, precipitation in June less than 80 mm, and precipitation in July less than 80 mm; Level 3: Meteorological conditions are not suitable for the occurrence of corn leaf blight, temperature <5℃ or temperature >35℃.

Images