METHOD AND DEVICE FOR RISK CONTROL IN THE ASSESSMENT OF VEHICLE ACCIDENT DAMAGE, ELECTRONIC DEVICE AND STORAGE MEDIUM

Abstract

Procedures for risk control in the assessment of vehicle accident damage, comprehensive: Recording a report on the assessment of vehicle accident damage from a current accident involving a vehicle, wherein the vehicle accident damage assessment report includes vehicle information and vehicle damage items relating to the current accident involving the vehicle; Calculating the vehicle damage items in the vehicle accident damage assessment report using a previously developed risk algorithm model in combination with the vehicle information to identify risk items that pose a risk from the vehicle damage items; and Intercepting risk positions or user indications of risk positions, whereby The previously created risk algorithm model includes a vehicle collision damage level algorithm model, the vehicle collision damage level algorithm model being created in the following steps: Manually marking collision points and damage levels of historical vehicle accident cases using historical reports on the assessment of vehicle accident damage of various vehicle types and in combination with damage photos from historical vehicle accident damage assessments, and Learning historical reports on the assessment of vehicle accident damage of different vehicle types, as well as profiles of collision points and damage levels, using a computer in a machine-based manner based on manually marked collision points and damage levels of historical vehicle accident cases, in order to obtain correlations between the collision points and damage levels and the vehicle damage positions, and whereby The calculation of the vehicle damage items in the vehicle accident damage assessment report using the vehicle collision point damage level algorithm model includes the following: Calculating the damage level of the collision point of the current accident based on the vehicle damage items in the recorded current report on the assessment of vehicle accident damage and the collision point, which is calculated by the vehicle collision point algorithm model in the risk algorithm model, thereby determining a damage limit of the parts in relation to the collision point and the damage level, in order to assess whether damaged parts include a damaged part that lies outside the damage limit, wherein, if the result of the assessment is that such a part is included, the damaged part that lies outside the damage limit is identified as a risk item. wherein the previously created risk algorithm model includes the vehicle collision point algorithm model, wherein the vehicle collision point algorithm model is created in the following steps: Virtual mapping of the vehicle into a corresponding diagram in a three-dimensional coordinate system according to the vehicle information and assignment of three-dimensional coordinates to vehicle parts according to the actual installation positions of the vehicle parts in the vehicle, Dividing the vehicle into different collision points and creating three-dimensional coordinate data for each of these collision points according to the vehicle information, thereby establishing relationships between the collision points and vehicle parts, where the calculation of the vehicle damage items in the report on the assessment of vehicle accident damage using the vehicle collision point algorithm model includes at least one of the following: Calculating a collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report for the vehicle, and comparing the current vehicle accident damage assessment report for the vehicle with historical vehicle accident damage assessment reports for the vehicle to calculate whether the same collision point or damaged part is present, whereby, if the result of the calculation is that it is present, the same collision point or damaged part is identified as a risk item; Assess whether, if it is calculated that there are multiple collision points of the current accident of the vehicle, there is a collision point among the collision points that does not conform to the collision logic, whereby, if the result of the assessment is that such a collision point exists, the collision point that does not conform to the collision logic is identified as a risk position.

Description

[0001] The present invention relates to the fields of vehicle information technology and vehicle damage assessment, in particular a method and a device for risk control in the assessment of vehicle accident damage, an electronic device and a storage medium.

[0002] With the rapid development of the Chinese economy and the steady increase in consumer spending, the vehicle has gradually transformed from a luxury item into an everyday means of transportation for families. Vehicle insurance, as a value-added service, has also developed rapidly, becoming the largest insurance sector with a 60% market share. Vehicle insurance is a mandatory insurance requirement imposed by the state on all motor vehicles. This insurance primarily serves to protect owners in the event of a traffic accident that could result in property damage or personal injury.

[0003] In the process of vehicle damage settlement, including damage assessment and determination, an accurate and appropriate evaluation of vehicle accident damage is a crucial aspect of insurance companies' obligations and the first key step in the claims settlement process. However, due to the skills and qualifications of the insurance employees responsible for assessing vehicle accident damage, as well as the complexity and specialized nature of such damage, there are often cases where the evaluation is inaccurate or inappropriate. This leads to losses for insurance companies and unnecessarily increases their compensation costs. Therefore, risk control in the evaluation of vehicle accident damage from the outset and the reduction of claims settlement costs are of paramount importance.

[0004] In the process of assessing vehicle accident damage at insurance companies, aimed at improving the speed and efficiency of claims settlement and increasing customer satisfaction, the process has gradually expanded from assessment by the insurance company's claims adjuster to assessment by repair shops, and even to self-assessment by the vehicle owner in the case of minor damage. However, due to profit motives, discrepancies in assessments by repair shops are inevitable; and self-assessment by the vehicle owner also carries the risk of inaccuracies due to a lack of expertise. Customer needs are constantly evolving, and business models must be transformed to meet these changing needs.Traditional empirical assessment and auditing methods are no longer adapted to the evolution of the industry, and efficient and intelligent risk control methods are lacking. Furthermore, the transition from manual to automated auditing is a clear trend.

[0005] The leaks in claims assessment in motor vehicle insurance are estimated by the industry to exceed ten billion yuan, significantly impacting the costs of claims settlement for insurance companies, affecting their solvency, and leading to a massive waste of public assets; at the same time, this is also a chronic problem in the insurance industry that has long been difficult to solve.

[0006] It is obvious that insurance companies face an urgent challenge in assessing vehicle accident damage, in order to carry out efficient and accurate assessments while simultaneously strengthening their competence in risk identification in order to plug the gaps in damage assessment.

[0007] From WO 2023 / 006974 A1, a method for the automatic detection and assessment of damage to a vehicle and the provision of estimated repair / replacement costs is disclosed. US 2021 / 0256616 A1 describes the determination of an insurance risk for motor vehicles based on past claims using artificial intelligence. DE 112012003110 T5 describes a method for the automatic verification of application data for maliciously manipulated data entries using machine learning.

[0008] In view of this, embodiments of the present invention provide a method and a device for risk control in the assessment of vehicle accident damage, which, based on artificial intelligence, industry big data standardization and big data mining analysis, can automatically warn of possible losses in the assessment of vehicle accident damage in advance, in order to implement risk control and ensure the accuracy of the assessment of vehicle accident damage.

[0009] To solve the above-mentioned problem, a method for risk control in the assessment of vehicle accident damage according to claim 1 is provided according to a first aspect of the present invention.

[0010] The risk control method for assessing vehicle accident damage according to the first aspect of the present invention can use a pre-defined risk algorithm model to intelligently evaluate the damage items of the vehicle involved that have been entered or reported into the preliminary information on assessing vehicle accident damage. This allows risk items that pose a risk to be calculated and either directly and necessarily intercepted or displayed to the user in a generated risk control report to alert them to these risk items (which are judged to be either unrelated to the current vehicle accident or inappropriate).This allows the user to take appropriate measures based on the risk positions displayed by the system to reduce the risk in the assessment of vehicle accident damage and to achieve a more accurate and reasonable valuation result. This enables efficient and intelligent risk control in the assessment of vehicle accident damage and significantly reduces leaks in insurance claims.

[0011] In the risk control method for assessing vehicle accident damage according to the first aspect of the present invention, various calculations and assessments are performed using a previously created risk algorithm model. That is, an artificial intelligence method is used to perform a risk assessment by computer, thus enabling automated verification of the vehicle accident damage assessment. Compared to traditional manual verification, which depends on the professional technical expertise of the workers, this method significantly reduces personnel costs, saves time, and increases accuracy and objectivity.

[0012] Preferably, in the above procedure for risk control in the assessment of vehicle accident damage, the vehicle information may include the following: a vehicle identification number (VIN), a vehicle type (insured vehicle / third-party vehicle), a cause of the accident (collision, etc.), a damage assessment method (repair assessment, etc.).

[0013] According to the above procedure for risk control in the assessment of vehicle accident damage, targeted risk control based on vehicle information can be carried out, which increases reliability.

[0014] Preferably, the above procedure for risk control in the assessment of vehicle accident damage includes the previously created risk algorithm model:

[0015] a vehicle part damage relationship measure model, wherein the vehicle part damage relationship measure model is created in the following steps:

[0016] Performing a big data mining analysis using historical reports on the assessment of vehicle accident damage from various vehicle types as sample data, calculating and storing a damage correlation measure between vehicle parts when the vehicle is damaged in an accident,

[0017] wherein the damage causation measure includes at least one of the following: a probability that if one vehicle part is damaged and thus becomes a damaged part, another vehicle part will also become a damaged part; and a probability that if one vehicle part is damaged and thus becomes a damaged part, the other vehicle part will not become a damaged part.

[0018] Preferably, in the above procedure for risk control in the assessment of vehicle accident damage, the calculation of the vehicle damage items in the vehicle accident damage assessment report using the vehicle part damage correlation model includes at least one of the following:

[0019] Identify, based on the vehicle damage items in the recorded current vehicle accident damage assessment report of the vehicle, in combination with the damage correlation measure, whether another part that is strongly related to a damaged part included in the vehicle damage items is also recorded as a damaged part in the vehicle damage items, whereby if the result of the identification is that it is not recorded, the damaged part is identified as a risk item;

[0020] where the strong correlation means that if one vehicle part is damaged and thus becomes a damaged part, the probability that another vehicle part will also become a damaged part is higher than a predetermined threshold.

[0021] Preferably, the above procedure for risk control in the assessment of vehicle accident damage includes the previously created risk algorithm model: a vehicle original equipment standard information model, wherein the vehicle original equipment standard equipment model is created by big data standardization of vehicle original equipment information, where the vehicle original equipment information includes at least one of the following items: vehicle original equipment data, part codes, quantities of parts per vehicle, relationships between assemblies and individual parts, indicative part prices and quantities of auxiliary materials.

[0022] Preferably, in the above procedure for risk control in the assessment of vehicle accident damage, the calculation of the vehicle damage items in the vehicle accident damage assessment report based on the vehicle original standard equipment information model includes the following:

[0023] Identifying original vehicle equipment information that corresponds to the vehicle, based on the vehicle information using the vehicle original equipment standard equipment information model, and comparing the information about the damaged parts in the vehicle damage items with the original vehicle equipment information, whereby, if the result of the comparison between the two is inconsistent, the damaged parts that are inconsistent are identified as risk items.

[0024] Preferably, the above procedure for risk control in the assessment of vehicle accident damage includes the previously created risk algorithm model: a vehicle part standard repair process model, wherein the vehicle part standard repair process model is created through big data standardization of technical repair data and vehicle part repair process standards.

[0025] Preferably, in the above procedure for risk control in the assessment of vehicle accident damage, the calculation of the vehicle damage items in the vehicle accident damage assessment report using the vehicle part standard repair process model includes the following:

[0026] Comparing repair information of the damaged parts in the vehicle damage positions with the technical repair data and the repair process standards based on the vehicle information using the vehicle part standard repair process model, whereby, if the result of the comparison of the two is inconsistent, the damaged parts that are inconsistent are identified as risk positions.

[0027] Within the framework of the risk control method for assessing vehicle accident damage according to the first aspect of the present invention, the previously created risk algorithm model, which serves as the basis for the risk control calculation, comprises models in at least five dimensions, namely the vehicle collision point algorithm model, the vehicle collision point damage degree algorithm model, the vehicle part damage correlation measure model, the vehicle original standard equipment information model, and the vehicle part standard repair process model. Based on these models, which comprehensively consider the collision points, the damage degrees, the damage correlation measure between parts, as well as the original standard equipment and repair processes, and through machine learning based on data from historical real-world cases and on the contents of vehicle accident damage assessment reports, it is possible toto identify the vehicle's collision points and damage levels; using data mining algorithms, the probability of no damage to parts can be calculated for different make models, different collision points, and different damage levels; by subjecting vehicle original equipment data, part data, technical repair data, repair process standards, etc., to big data standardization, big data rules for risk identification and assessment in damage appraisal can be created, making it possible to carry out comprehensive, reliable, accurate, and intelligent risk control, thus improving risk control capabilities.

[0028] Furthermore, according to the first aspect of the present invention, the above method for risk control in the assessment of vehicle accident damage provides the industry with an efficient, intelligent, and convenient tool for risk control in the assessment of vehicle accident damage; it breaks with the patterns and concepts of traditional risk control in the assessment of vehicle accident damage, introduces technological innovations, and switches from traditional empirical rules to innovative big data mining analysis; the transition from manual to machine review significantly improves review efficiency, saves the insurance industry's claims settlement costs, and avoids the waste of public assets; based on artificial intelligence, industry big data standardization, and big data mining analysis, a complete risk control system is formed that addresses the industry's pain points.

[0029] Preferably, in the above risk control procedure for assessing vehicle accident damage, the user selects, based on the user's indication of the risk, whether a risk correction should be made based on the description of the individual risk items or not, wherein, if the risk correction is selected, a risk identification and assessment is carried out again using the risk algorithm model.

[0030] According to the above procedure for risk control in the assessment of vehicle accident damage, it is possible to perform a calibration and cyclical control of risk positions based on the user's selection, which increases usability and accuracy.

[0031] The second aspect of the present invention provides a device for risk control in the assessment of vehicle accident damage according to claim 11.

[0032] The device for risk control in the assessment of vehicle accident damage according to the second aspect of the present invention can achieve similar technical effects as in the first aspect.

[0033] The third aspect of the present invention provides an electronic device comprising one or more processors and a storage device for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, this results in the one or more processors implementing a method as described in the first aspect.

[0034] The fourth aspect of the present invention provides for a computer-readable medium on which a computer program is stored, wherein the program, when executed by a processor, implements a method as described in the first aspect.

[0035] The second to fourth aspects of the present invention are capable of achieving the same technical effects as the first aspect.

[0036] The further effects of the aforementioned unconventional optional designs will be explained below in conjunction with the specific examples of implementation. Brief description of the drawing

[0037] The accompanying drawings serve to better understand the present invention and do not constitute an inadmissible limitation of the invention. They show: Fig. 1 a schematic diagram of a main sequence of a method for risk control in the assessment of vehicle accident damage according to embodiments of the present invention; Fig. 2 a schematic diagram of steps for creating a vehicle collision point algorithm model; Fig. 3 a schematic diagram of steps for creating a vehicle collision site damage level algorithm model; Fig. 4 a schematic diagram of the rules for classifying risk positions according to their risk level; Fig. 5 a schematic diagram of a further preferred sequence of the procedure for risk control in the assessment of vehicle accident damage according to embodiments of the present invention; Fig. 6 an exemplary diagram of a risk control report from an example of the risk control procedure for assessing vehicle accident damage according to embodiments of the present invention; Fig. 7 a schematic diagram of a process from an example of the method for risk control in the assessment of vehicle accident damage according to embodiments of the present invention; Fig. 8 a schematic diagram of a main configuration of a device for risk control in the assessment of vehicle accident damage according to embodiments of the present invention; Fig. 9 a diagram of an exemplary system architecture to which embodiments of the present invention can be applied; Fig. 10 a schematic diagram of the architecture of a computer system suitable for use in the implementation of an end device or a server according to embodiments of the present invention. Detailed descriptions

[0038] The following are exemplary embodiments of the present invention, which include various details of the embodiments of the invention to facilitate understanding, with reference to the accompanying drawings; these should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Likewise, for the sake of clarity and conciseness, the description of known functions and structures is omitted in the following description.

[0039] The vehicles concerned in the procedure for risk control in the assessment of vehicle accident damage according to the present application relate in particular to, but not exclusively to, four-wheeled passenger cars.

[0040] Fig. Figure 1 shows a schematic diagram of a main sequence of a risk control method for assessing vehicle accident damage according to embodiments of the present invention. As in Fig. As shown in Figure 1, the method for risk control in the assessment of vehicle accident damage according to embodiments of the present invention mainly comprises steps S101 to S103.

[0041] In step S101, the system can generate a report on the assessment of vehicle accident damage for a current accident involving a vehicle. This report includes vehicle information and damage items related to the accident. The vehicle information can include the make and model, the Vehicle Identification Number (VIN), the vehicle type (e.g., insured vehicle / third-party vehicle), the cause of the accident (e.g., collision), and the damage assessment method (e.g., repair assessment). The damage items can include information about damaged parts and information about the repair processes for those parts.

[0042] Furthermore, it should be noted that the system's process of recording the vehicle accident damage assessment report for the current accident involves at least two scenarios. The first scenario is that the user manually enters or types the required information into the system, and the system then records the vehicle accident damage assessment information. The second scenario is that a third party (for example, a repair shop) has already prepared a vehicle accident damage assessment report, and the system records this report by integrating it into the system. Therefore, the risk control procedure for vehicle accident damage assessment in this example exhibits a high degree of general applicability.It is suitable not only for the scenario in which the information on the assessment of vehicle accident damage is directly captured by the system, but also for the scenario in which the system captures the information on vehicle accident damage already established by the third party and performs a calculation for risk assessment.

[0043] In step S102, the system can calculate the vehicle damage items in the vehicle accident damage assessment report using a previously created risk algorithm model in combination with the vehicle information to identify risk items that pose a risk. These risk items are those that are deemed either unrelated to the current vehicle accident or inappropriate.

[0044] In step S103, the system can intercept risk positions or alert the user to them. More specifically, the system can generate a risk control report, which is previewed for the user. The risk control report can list the risk positions of the alert class, including risk position names, rule labels (such as risk reasons, etc.), rule classifications (such as fraud prevention, loss logic, etc.), recommended actions, reducible amounts, comments, and so on.

[0045] The risk control method for assessing vehicle accident damage according to the first aspect of the present invention, comprising steps S101 to S103, can utilize a pre-defined risk algorithm model to intelligently evaluate the damage items of the vehicle involved that have been entered or reported into the preliminary information on assessing vehicle accident damage. This allows for the calculation of risk items that pose a risk. These risk items can then be, for example, directly and mandatorily intercepted, or a risk control report can be generated and displayed to the user to alert them to these risk items. Based on the risk items displayed by the system, the user can take appropriate measures to reduce the risk in assessing vehicle accident damage and to achieve a more accurate and reasonable assessment result.This enables efficient and intelligent risk control in the assessment of vehicle accident damage and significantly reduces leaks at the insurance company. Furthermore, various calculations and assessments are performed using a pre-developed risk algorithm model; that is, an artificial intelligence method is used to conduct risk assessments automatically, thus enabling automated verification of vehicle accident damage assessments. Compared to traditional manual verification, which relies on the professional technical expertise of the workers, personnel costs are significantly reduced, time is saved, and accuracy and objectivity are increased.

[0046] According to embodiments of the present invention, the risk algorithm model previously created in step S102 may comprise: a vehicle collision point algorithm model for calculating a collision point of the current accident of the vehicle and identifying the risk positions based on the calculation result; a vehicle collision point damage degree algorithm model for calculating a damage degree of the collision point and identifying the risk positions based on the calculation result; a vehicle part damage correlation measure model for identifying the risk positions that represent risk in the vehicle damage positions according to the damage correlation measure between vehicle parts;A vehicle original equipment standard information model to identify whether the information about the damaged parts matches the vehicle parts' original equipment information and the vehicle parts' original standard information in order to identify the risk positions; and a vehicle part standard repair process model to calculate whether the repair information of the damaged parts matches the technical repair data and the repair process standards in order to identify the risk positions.

[0047] It should be noted that while the risk algorithm model presented above comprises five models—namely a vehicle collision point algorithm model, a vehicle collision point damage degree algorithm model, a vehicle part damage correlation measure model, a vehicle original equipment standard equipment information model, and a vehicle part standard repair process model—this does not represent a limitation. For example, it is sufficient for the risk algorithm model to include at least one of these models.

[0048] The vehicle collision point algorithm model of this embodiment is created beforehand in the following steps: mapping the vehicle virtually into a corresponding diagram in the three-dimensional coordinate system according to the vehicle information and assigning three-dimensional coordinates to vehicle parts according to the actual installation positions of the vehicle parts in the vehicle (according to S2011 in Fig. 3); Dividing the vehicle into different collision points and creating three-dimensional coordinate data for each of these collision points according to the vehicle information, thereby establishing relationships between the collision points and vehicle parts (according to S2012 in Fig. 3).

[0049] More precisely, 3D coordinate gyroscope data is generated for the installation location of the vehicle parts. According to the different body styles of the vehicle (mainly divided into four-door sedan, two-door sedan, five-door hatchback, SUV, MPV and van, a total of six types), the vehicle is virtually mapped into a corresponding three-dimensional diagram, such as a cuboid, and a three-dimensional coordinate system is created for the installation location of the parts.For example, if a car is a four-door sedan, the three-dimensional coordinate system can be created as follows: Coordinates on the x-axis (from the left to the right side of the vehicle) = (1, 2, 3, 4, 5, 6, 7, 8, 9); Coordinates on the y-axis (from the front to the back of the vehicle) = (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12); Coordinates on the z-axis (from the ground to the roof of the vehicle) = (1, 2, 3); the three-dimensional coordinates of the parts' installation location are determined according to the actual installation location of the parts in the vehicle. For example, for the "front headlight assembly (left)" in a car, the coordinate values ​​could be (1, 1, 2) or (2, 1, 2).

[0050] Furthermore, the vehicle's collision points are identified and the three-dimensional coordinate ranges for these points are defined. The exterior of the vehicle body is divided into a total of 10 collision points: front, left front, right front, left center, right center, rear front, left rear, right rear, roof section, and underbody. Three-dimensional coordinate data is generated for each collision point according to the body shape. This allows the names of the parts at the various collision points to be calculated based on the body shape, which can be determined by analyzing the vehicle identification number (VIN), and the actual coordinates of the parts' installation locations.

[0051] The vehicle collision damage assessment algorithm model can be obtained as follows: Using historical reports on the assessment of vehicle accident damage for various vehicle types as sample data, a computer performs machine learning to establish correlations between the damage levels of the collision points and the vehicle damage locations. More specifically, the model can, for example, be implemented in the Fig. The 3 steps shown are created: marking collision points and damage levels of historical vehicle accident cases manually using historical reports on the assessment of vehicle accident damage and in combination with damage photos from historical assessments of vehicle accident damage, as in step S3011 in Fig. 3 shown; learning the historical reports on the assessment of vehicle accident damage of the different vehicle types as well as profiles of collision points and damage levels by computer in a machine-like manner based on the manually marked collision points and damage levels of the historical vehicle accident cases in order to obtain relationships between the collision points and damage levels and the vehicle damage positions, as shown in step 3012.

[0052] In this embodiment, vehicle collision damage grades can be classified as follows: minor collision damage, light collision damage, moderate collision damage, severe collision damage, and extremely severe collision damage, for a total of five categories. For example, as mentioned earlier, the collision locations and damage grades are manually marked in the historical vehicle accident damage assessment reports using damage photographs from those reports. Based on these manually marked collision locations and damage grades, the computer automatically learns the historical vehicle accident damage assessment reports for the various vehicle types, as well as the profiles of the different collision locations and damage grades.Thus, it is possible to learn the relationships between collision points and damage levels and vehicle damage positions, and to create a vehicle collision point damage level algorithm model. For example, the computer can, in the case that is in . Fig. Figure 4 illustrates how, based on the vehicle collision point damage grade algorithm model, if the vehicle accident damage assessment report contains damaged parts such as the front headlight assembly (right), the front bumper cover, and the front fender (right), the relationships between these parts allow for precise identification of the accident as a "minor front-right collision." Furthermore, the probability can be simultaneously calculated that no damage items will occur for different vehicle types (e.g., FAW Audi, 2022 A6L), different body styles (e.g., four-door sedan), different collision points, and different collision damage grades.

[0053] Furthermore, using the vehicle collision damage severity algorithm model, a damage limit for the parts can be determined in relation to the collision point and the damage severity after calculating the damage severity of the collision point in the current accident. For example, in the case of minor collision damage to the front, it can be determined based on big data mining of historical cases that the damage limit in this case should be limited to external parts and should not include internal parts. If damage to the engine assembly is found in the vehicle accident damage assessment report, this does not correspond to the defined damage limit. Damage to the engine assembly does not fall into the category of minor collision damage to the front.

[0054] As mentioned above, using the vehicle collision point algorithm model, it is possible to perform at least one of the following calculations on the vehicle damage items in the vehicle accident damage assessment report: calculating a collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report of the vehicle, and comparing the current vehicle accident damage assessment report of the vehicle with historical vehicle accident damage assessment reports of the vehicle to calculate whether the same collision point or damaged part is present, whereby, if the result of the calculation is that it is present, the same collision point or damaged part is identified as a risk item;Assess whether, if it is calculated that there are multiple collision points of the current accident of the vehicle, there is one collision point among the collision points that is inconsistent with the collision logic, and if the result of the assessment is that such a collision point exists, the collision point that is inconsistent with the collision logic is identified as a risk position.

[0055] As mentioned above, using the vehicle collision point damage level algorithm model, it is possible to calculate a damage level of the collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report and the collision point calculated by the vehicle collision point algorithm model. This determines a damage limit of the parts in relation to the collision point and the damage level, in order to assess whether the damaged parts include a damaged part that lies outside the damage limit. If the result of the assessment is that such a part is included, the damaged part that lies outside the damage limit is identified as a risk item.

[0056] In particular, the collision point that does not conform to the collision logic refers to the collision point that is not adjacent to other collision points, or to the discrete collision point. For example, if the vehicle collision point algorithm model is used to determine the collision points of a vehicle accident at the front, left front, and rear, then the non-adjacent or discrete collision point at the rear represents the collision point that does not conform to the collision logic.

[0057] The in Fig. The vehicle part damage correlation measure model shown can be created as follows: Perform a big data mining analysis using historical reports on the assessment of vehicle accident damage for various vehicle types as sample data. Calculate and store a damage correlation measure between vehicle parts when the vehicle is damaged in an accident. This damage correlation measure includes at least one of the following: a probability that if one vehicle part is damaged and thus becomes a damaged part, another vehicle part will also become a damaged part.

[0058] Based on historical vehicle accident damage assessment reports, a big data mining calculation of the parts accident damage correlation measure is performed to determine the relationship between two parts, A and B, of the same brand model in accident damage. This correlation includes a confidence level, which describes the probability that damage to part A will simultaneously result in damage to part B; a lift level, which measures the independence of damage between parts A and B; and a conviction level, which measures the probability of a prediction error and is also used to assess the independence of damage between parts A and B. Therefore, based on the vehicle part damage correlation measure model, risk identification for damage items in the vehicle accident damage assessment report can be performed.For example, if it is calculated that there is a strong correlation between the two parts A and B, and it is identified that the report shows damage to part A but no damage to part B, then part A will be assessed as a risk position.

[0059] The strong correlation means that if one vehicle part is damaged and thus becomes a damaged part, the probability that another vehicle part will also become a damaged part is higher than a predetermined threshold (90%).

[0060] The above method for risk control in the assessment of vehicle accident damage according to this embodiment is able to create and perfect a vehicle collision point damage degree algorithm model and a vehicle part damage correlation measure model on the basis of historical real data accurately, reliably and intelligently, as well as to carry out risk identification and assessment.

[0061] Furthermore, the vehicle original equipment standard information model and the vehicle part standard repair process model, which are in Fig. 2 are shown, as described below.

[0062] The vehicle original equipment information model is created, for example, through the big data standardization of vehicle original equipment information and is related to vehicle part original equipment information and vehicle part original standard information. The vehicle part original equipment information and the vehicle part original standard information include big data such as vehicle original equipment data, part codes, quantities of parts per vehicle, relationships between assemblies and individual parts, target prices of parts, and quantities of consumables such as engine oil.

[0063] Each vehicle has a unique Vehicle Identification Number (VIN) upon delivery. Each vehicle has its own original equipment standard corresponding to its VIN, such as whether the headlights are halogen or xenon, with or without adaptive headlights, etc. Different equipment configurations for the same model result in significant price differences for parts. The system is able to accurately analyze the VIN and identify the vehicle's equipment standard. Simultaneously, it can also identify the corresponding equipment standards for the parts in the vehicle accident damage assessment report. Both are then subjected to risk identification and assessment to identify and evaluate risks that deviate from the original equipment standard, thereby mitigating those risks.

[0064] The vehicle, as defined by its vehicle identification number (VIN), also has standards for the quantities of parts per vehicle, the relationships between assemblies and individual parts, and so on. When comparing and identifying the damaged parts in the vehicle accident damage assessment report, it is checked whether there are any instances of parts quantities being exceeded or simultaneous damage to assemblies and individual parts, in order to mitigate risk. For example, the original manufacturer specifies a quantity of "1" per vehicle for the "front headlight assembly (left)." If the report indicates a quantity of "2," this is an overage. Furthermore, the "front headlight assembly (left)" already includes a "front headlight housing (left)" as a separate part.If the report indicates simultaneous damage to the "front headlight assembly (left)" and the "front headlight housing (left)", risk mitigation should be carried out.

[0065] The vehicle part standard repair process model is created through the big data standardization of technical repair data and vehicle part repair process standards. Vehicle parts all have standardized repair processes. For example, the "front bumper cover" cannot be repaired simultaneously by "replacement" and "repair." "Replacement" and "repair" are two mutually exclusive repair processes. Therefore, the vehicle part standard repair process model is used to check for instances of mutually exclusive repair processes in the report, and so on. If such instances are found, risk mitigation is implemented.

[0066] Fig. Figure 4 shows an example diagram of the rules for the risk positions that can be calculated by the vehicle collision point algorithm model, the vehicle collision point damage level algorithm model, the vehicle part damage correlation model, the vehicle original equipment information model, and the vehicle part standard repair process model. As shown, in this embodiment, the risk positions are divided into a class of risk interception rules and a class of risk alert rules, the latter with a risk level below that of the risk interception rules.Generally speaking (not restrictively), the risk positions calculated by the vehicle part standard repair process model and the vehicle original equipment standard information model can be classified as risk interception rules and forcibly intercepted, so that the risk positions assigned to the risk interception rules are removed or modified from the vehicle accident damage assessment information; the risk positions calculated by the vehicle collision point algorithm model, the vehicle collision point damage grade algorithm model, and the vehicle part damage correlation measure model can be classified as risk alert rules and intercepted or allowed to pass based on subsequent user actions.

[0067] As in Fig. As shown in Figure 4, the class of risk interception rules mainly refers to positions 1 to 9. They have a particularly high risk level and are sufficient to qualify as valuation risks; the class of risk indicator rules mainly refers to positions 10 to 14. Their risk level is lower compared to the class of risk interception rules, belongs to the highly suspicious risks, and requires the claims adjuster to re-verify the authenticity and accuracy of the loss and provide sufficient justification for the valuation. Furthermore, the far right column of Fig. 4 explains the basis of the corresponding risk rules. The in Fig. The explanations shown in section 4 can, for example, be presented accordingly in the risk control report.

[0068] In the above procedure for risk control in the assessment of vehicle accident damage according to this embodiment, the risk items are assigned risk levels and controlled in different ways. Compared to the user subsequently processing all risk items, this can reduce the operational workload for the user and also effectively help to directly filter out risk items that represent a high risk.

[0069] Within the framework of the risk control method for assessing vehicle accident damage according to the first aspect of the present invention, the previously created risk algorithm model, which serves as the basis for the risk control calculation, comprises models in at least five dimensions, namely the vehicle collision point algorithm model, the vehicle collision point damage degree algorithm model, the vehicle part damage correlation measure model, the vehicle original standard equipment information model, and the vehicle part standard repair process model. Based on these models, which comprehensively consider the collision points, the damage degrees, the damage correlation measure between parts, as well as the original standard equipment and repair processes, and through machine learning based on data from historical real-world cases and on the contents of vehicle accident damage assessment reports, it is possible toto identify the vehicle's collision points and damage levels; using data mining algorithms, the probability of no damage to parts can be calculated for different make models, different collision points, and different damage levels; by subjecting vehicle original equipment data, part data, technical repair data, repair process standards, etc., to big data standardization, big data rules for risk identification and assessment in damage appraisal can be created, making it possible to carry out comprehensive, reliable, accurate, and intelligent risk control, thus improving risk control capabilities.

[0070] Furthermore, according to the first aspect of the present invention, the above method for risk control in the assessment of vehicle accident damage provides the industry with an efficient, intelligent, and convenient tool for risk control in the assessment of vehicle accident damage; it breaks with the patterns and concepts of traditional risk control in the assessment of vehicle accident damage, introduces technological innovations, and switches from traditional empirical rules to innovative big data mining analysis; the transition from manual to machine review significantly improves review efficiency, saves the insurance industry's claims settlement costs, and avoids the waste of public assets; based on artificial intelligence, industry big data standardization, and big data mining analysis, a complete risk control system is formed that addresses the industry's pain points.

[0071] Furthermore, as in Fig. As shown in Figure 5, in the procedure for risk control in the assessment of vehicle accident damage according to this embodiment, the user selects after step S103 whether a risk correction should be made based on the description of the individual risk positions in the risk control report or not, wherein, if the risk correction is selected, a risk identification and assessment is carried out again using the risk algorithm model to enable calibration and cyclical control of the risk positions, which increases usability and accuracy.

[0072] The basic idea of ​​the risk control method for assessing vehicle accident damage according to embodiments of the present invention has been described above. The following section refers to the Fig. 6 to Fig. 7. A clearer description is given using an example to facilitate understanding. However, experts should understand that this example is not restrictive, but merely an illustration.

[0073] First, the user can request the creation of a vehicle accident damage assessment order within the system. Creating a vehicle accident damage assessment order involves the following process: entering basic vehicle information and generating a damage assessment task. In this example, the user (assessor) accesses the main page of an app via a mobile device, such as a smartphone, or the homepage of a website via a desktop computer. Using OCR (Optical Character Recognition) or manual entry of a vehicle registration number and vehicle identification number (VIN), the system analyzes the VIN to identify the vehicle model, confirm the vehicle's identity, and retrieve vehicle information. The user then selects a repair shop (4S shop, garage, etc.).) selects and completes the basic information on the valuation prices in order to prepare a preliminary report on the valuation of vehicle accident damage (S1).

[0074] After the system receives a request to generate a vehicle accident damage assessment report, the user is shown an input page where they can enter vehicle damage items (S2), such as the names, quantities, repair processes (replacement, disassembly, painting, repair, low-carbon repair, etc.) of the damaged parts, and so on. For example, in this example, the user enters "Replacement of right front headlight, air filter, front bumper, painting of right front fender" into the input field. The system then performs an intelligent semantic analysis and automatically recognizes and selects: front headlight assembly (right) [part] [replacement], air filter assembly [part] [replacement], front bumper trim [part] [replacement], front fender (right) [paint]. This determines the vehicle damage items.At the same time, the original part codes, quantities per vehicle, notes, manufacturer's recommended guide prices and system reference prices are automatically adopted.

[0075] Subsequently, the assessor individually confirms the damage amounts based on the reference prices for the damage items suggested by the system, in order to finally determine the vehicle damage items and total damage amounts, and thus to prepare a complete report on the assessment of vehicle accident damage (S3).

[0076] The system then receives the vehicle accident damage assessment report and performs a corresponding risk identification calculation for the damage assessment based on the information contained about the vehicle and the damaged parts using the vehicle collision point algorithm model, the vehicle collision point damage degree algorithm model, the vehicle part damage correlation measure model, the vehicle original standard equipment information model, and the vehicle part standard repair process model, so that, based on these risk algorithm models, a risk identification and assessment is carried out for the contents of the vehicle accident damage assessment report in order to then calculate risk positions, recommendations for action, and reduced amounts (S4).

[0077] After the calculation, the system automatically generates a risk control report (S5). The report includes names of the risk positions, risk descriptions, rule classifications, recommendations for action, reducible amounts, and comments (see Fig. 6) For risks subject to interception rules, the system can perform a forced interception as early as the input phase of the loss items. For risks subject to advisory rules, assessors should be able to re-examine and confirm based on the risk control report, correct an erroneous assessment, or provide sufficient reasons for the assessment (S6).

[0078] In the technical concept of the present invention, risk control can be performed automatically using artificial intelligence. One of the core aspects of the artificial intelligence is the modeling of damage patterns in vehicle collisions, which is achieved through machine learning based on extensive historical data and big data mining analysis to calculate the relationship between the vehicle collision damage model and the detailed damage report.Specifically, machine learning is performed based on the content of vehicle accident damage assessment reports to identify the collision points and damage levels of the vehicle; data mining algorithms are used to calculate the probability that no damage will occur to the parts for different make models, different collision points, and different damage levels; and the vehicle original equipment data, part data, technical repair data, and repair process standards are subjected to big data standardization to create big data rules for risk identification and assessment in damage evaluation.

[0079] The risk control method for assessing vehicle accident damage according to the present invention provides an efficient, intelligent, and convenient tool for risk control in the assessment of vehicle accident damage; it breaks with the patterns and concepts of traditional risk control of vehicle accident damage and moves from traditional empirical rules to innovative big data mining analysis; the transition from manual to machine review significantly improves review efficiency, saves the insurance industry's claims settlement costs, and avoids the waste of public assets; based on artificial intelligence, industry big data standardization, and big data mining analysis, a complete risk control system is formed that addresses the industry's pain points.

[0080] The embodiments of the present invention also provide a device for risk control in the assessment of vehicle accident damage, as described in Fig. Figure 8 shows the device 200 for risk control in the assessment of vehicle accident damage, comprising a data acquisition unit 201 for capturing a report on the assessment of vehicle accident damage of a current accident of a vehicle, wherein the report on the assessment of vehicle accident damage contains vehicle information of the vehicle and vehicle damage items relating to the current accident of the vehicle; a risk identification and assessment unit 202 for calculating the vehicle damage items in the report on the assessment of vehicle accident damage using a previously created risk algorithm model in combination with the vehicle information, in order to identify risk items that pose a risk from the vehicle damage items, wherein the risk items are assessed as not belonging to the current vehicle accident or as inappropriate;and a risk processing unit 203 for intercepting risk positions or user indications of risk positions.

[0081] The device 200 for risk control in the assessment of vehicle accident damage according to the embodiment of the present invention can achieve similar technical effects to the method for risk control in the assessment of vehicle accident damage in the first aspect. Although not shown here, the device 200 for risk control in the assessment of vehicle accidents is also capable of performing various processes of the method for risk control in the assessment of vehicle accident damage in the first aspect.

[0082] Fig. Figure 9 shows a diagram of an exemplary system architecture 300 to which the method and device for risk control in the assessment of vehicle accident damage can be applied.

[0083] As in Fig. As shown in Figure 9, the system architecture can include 300 end devices (301, 302, 303), a network (304), and a server (305). The network (304) serves as the medium for providing a communication link between the end devices (301, 302, 303) and the server (305). The network (304) can include various connection types, such as wired, wireless, or fiber optic connections.

[0084] The user can use terminals 301, 302, and 303 to interact with server 305 over network 304 to receive or send messages, etc. Various communication client applications can be installed on terminals 301, 302, and 303, such as a damage assessment application, a web browser application, a search application, an instant messaging tool, etc. (examples only).

[0085] The terminal devices 301, 302, 303 can be various electronic devices that include a display and support browsing the Internet, including but not limited to smartphones, tablets, laptops, desktop computers and the like.

[0086] Server 305 can be a server providing various services, such as a backend management server that offers data support for the vehicle identification number (VIN) entered by the user via terminals 301, 302, and 303 (this is just an example). The backend management server can analyze and process data such as received product requests and report the processing results (for example, vehicle make and model information) back to the terminals.

[0087] It should be noted that the risk control procedure for assessing vehicle accident damage, which is provided in embodiments of the present invention, is generally executed by the server 305, and consequently the risk control device for assessing vehicle accident damage is normally provided in the server 305.

[0088] It should be understood that the number of end devices, networks, and servers in Fig. Figure 9 is only an example. Depending on the implementation requirements, any number of end devices, networks, and servers may be present.

[0089] The following refers to Fig. Figure 10 describes a schematic diagram of the architecture of a computer system 400 suitable for use in the implementation of an end device or a server according to embodiments of the present invention. The diagram in Fig. The terminal device shown in Figure 10 is merely an example and is not intended to represent any limitations of the functionality and scope of application of the embodiments of the present invention.

[0090] As in Fig. As shown in Figure 10, the Computer System 400 comprises a central processing unit (CPU) 401, which can perform various suitable actions and processes based on programs stored in a read-only memory (ROM) 402 or loaded from a memory section 408 into a random-access memory (RAM) 403. Various programs and data required for the operation of the System 400 are also stored in the RAM 403. The CPU 401, the ROM 402, and the RAM 403 are interconnected via the bus 404. An input / output interface (I / O) 405 is also connected to the bus 404.

[0091] The following components are connected to the I / O interface 405: an input unit 406, which includes a keyboard, mouse, and similar devices; an output unit 407, which includes a cathode ray tube (CRT), liquid crystal display (LCD), and similar devices, as well as a speaker and similar devices; a storage unit 408, which includes a hard disk and similar devices; and a communication unit 409, which includes a network interface card such as a LAN card, a modem, and similar devices. The communication unit 409 handles communication processing over a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. Removable media 411, such as floppy disks, CDs, magneto-optical disks, semiconductor memory, etc., are installed in the drive 410 as needed to install computer programs read from them into the storage unit 408 when required.

[0092] In particular, the method described in the flowcharts mentioned above can be implemented as a computer software program according to embodiments of the present invention. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, wherein the computer program contains program code for executing the method depicted in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network via the communication part 409 and / or installed from the removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, the functions defined in the system of the present invention are performed.

[0093] It should be noted that the computer-readable medium presented within the scope of this invention can be either a computer-readable signal storage medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can, for example, but not exclusively, be a system, device, or instrument that is electrical, magnetic, optical, electromagnetic, infrared, or semiconducting, or any combination thereof.More specific examples of computer-readable storage media include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, random-access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), a fiber optic cable, a portable compact read-only storage device (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium can be any physical medium that contains or stores a program, wherein the program can be used or combined with an instruction execution system, instruction execution device, or instruction execution apparatus.In the present invention, the computer-readable signaling medium can comprise data signals transmitted in the baseband or as part of a carrier signal, wherein these data signals carry the computer-readable program code. These transmitted data signals can take various forms, including, but not limited to, electromagnetic signals, light signals, or any suitable combination thereof. The computer-readable signaling medium can also be any computer-readable medium other than the computer-readable storage medium capable of sending, distributing, or transmitting programs that can be used or combined with an instruction execution system, instruction execution device, or instruction execution apparatus. The program code contained in the computer-readable medium can be transmitted via any suitable medium, including, but not limited to: wirelessly, electrically, optically, RF, etc.or any suitable combination thereof.

[0094] The flowcharts and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, a program section, or a subcode, wherein this subcode contains one or more executable instructions that serve to implement specified logical functions. It should also be noted that in some alternative embodiments, the functions marked in the boxes may occur in a different order than shown in the drawings. For example, two consecutive boxes may actually be executed largely in parallel, and sometimes they may be executed in reverse order, depending on the function involved.It should also be noted that each box in the block diagram or flowchart, and combinations of boxes in the block diagrams or flowcharts, can be implemented by a dedicated, hardware-based system that performs the specified functions or actions, or by a combination of dedicated hardware and computer instructions.

[0095] The units described in the exemplary embodiment of the present invention can be implemented using either software or hardware. The described units can also be integrated into a processor, for example, a processor comprising a detection unit, a risk identification and assessment unit, and a risk processing unit. In certain cases, the designations of these units do not necessarily restrict the units themselves. For example, the risk identification and assessment unit can also be described as a unit that queries a connected server to perform risk identification and assessment for vehicle accident damage.

[0096] As a further aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the exemplary embodiment above; it may also exist separately, without being incorporated into this device. The computer-readable medium carries one or more programs which, when executed by the device, cause the device to: capture a report on the assessment of vehicle accident damage to a vehicle, wherein the report on the assessment of vehicle accident damage contains vehicle information and information on vehicle damage items relating to the vehicle;Calculating the information of the vehicle damage items in the vehicle accident damage assessment report using a previously created risk algorithm model in combination with the vehicle information, in order to assess risk items that pose a risk from the information of the vehicle damage items in the vehicle accident damage assessment report, whereby the risk items are judged as not belonging to the current vehicle accident or as inappropriate; and intercepting the risk items or generating a risk control report that is displayed to the user in a preview.

[0097] The present invention breaks with the patterns and concepts of traditional risk control of vehicle accident damage and switches from traditional empirical rules to innovative big data mining analysis, provides an efficient, intelligent and convenient tool for risk control of vehicle accident damage, saves the costs of claims settlement for the insurance industry and avoids the waste of social assets.

[0098] The specific embodiments described above do not constitute a limitation of the scope of protection of this invention. A person skilled in the art should understand that, depending on design requirements and other factors, various modifications, combinations, sub-combinations, and alternatives are possible.

Claims

[1] Procedures for risk control in the assessment of vehicle accident damage, including: Recording a report on the assessment of vehicle accident damage from a current accident involving a vehicle, wherein the vehicle accident damage assessment report includes vehicle information and vehicle damage items relating to the current accident involving the vehicle; Calculating the vehicle damage items in the vehicle accident damage assessment report using a previously developed risk algorithm model in combination with the vehicle information to identify risk items that pose a risk from the vehicle damage items; and Intercepting risk positions or user indications of risk positions, whereby The previously created risk algorithm model includes a vehicle collision damage level algorithm model, the vehicle collision damage level algorithm model being created in the following steps: Manually marking collision points and damage levels of historical vehicle accident cases using historical reports on the assessment of vehicle accident damage of various vehicle types and in combination with damage photos from historical vehicle accident damage assessments, and Learning historical reports on the assessment of vehicle accident damage of different vehicle types, as well as profiles of collision points and damage levels, using a computer in a machine-based manner based on manually marked collision points and damage levels of historical vehicle accident cases, in order to obtain correlations between the collision points and damage levels and the vehicle damage positions, and whereby The calculation of the vehicle damage items in the vehicle accident damage assessment report using the vehicle collision point damage level algorithm model includes the following: Calculating the damage level of the collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report and the collision point, calculated by the vehicle collision point algorithm model in the risk algorithm model, thereby determining a damage limit of the parts in relation to the collision point and the damage level, in order to assess whether damaged parts include a damaged part that lies outside the damage limit, wherein, if the result of the assessment is that such a part is included, the damaged part that lies outside the damage limit is identified as a risk item. wherein the previously created risk algorithm model includes the vehicle collision point algorithm model, wherein the vehicle collision point algorithm model is created in the following steps: Virtual mapping of the vehicle into a corresponding diagram in a three-dimensional coordinate system according to the vehicle information and assignment of three-dimensional coordinates to vehicle parts according to the actual installation positions of the vehicle parts in the vehicle, Dividing the vehicle into different collision points and creating three-dimensional coordinate data for each of these collision points according to the vehicle information, thereby establishing relationships between the collision points and vehicle parts, where the calculation of the vehicle damage items in the report on the assessment of vehicle accident damage using the vehicle collision point algorithm model includes at least one of the following: Calculating a collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report for the vehicle, and comparing the current vehicle accident damage assessment report for the vehicle with historical vehicle accident damage assessment reports for the vehicle to calculate whether the same collision point or damaged part is present, whereby, if the result of the calculation is that it is present, the same collision point or damaged part is identified as a risk item; Assess whether, if it is calculated that there are multiple collision points of the current accident of the vehicle, there is a collision point among the collision points that does not conform to the collision logic, whereby, if the result of the assessment is that such a collision point exists, the collision point that does not conform to the collision logic is identified as a risk position. [2] Method for risk control in the assessment of vehicle accident damage according to claim 1, characterized by , that the previously created risk algorithm model includes a vehicle part damage relationship measure model, wherein the vehicle part damage relationship measure model is created in the following steps: Performing a big data mining analysis using historical reports on the assessment of vehicle accident damage from various vehicle types as sample data, calculating and storing a damage correlation measure between vehicle parts when the vehicle is damaged in an accident, wherein the damage causation measure includes at least one of the following: a probability that if one vehicle part is damaged and thus becomes a damaged part, another vehicle part will also become a damaged part; and a probability that if one vehicle part is damaged and thus becomes a damaged part, the other vehicle part will not become a damaged part. [3] Method for risk control in the assessment of vehicle accident damage according to claim 2, characterized by, that the calculation of vehicle damage items in the report on the assessment of vehicle accident damage using the vehicle part damage correlation measure model includes at least one of the following: Identify, based on the vehicle damage items in the recorded current vehicle accident damage assessment report of the vehicle, in combination with the damage correlation measure, whether another part that is strongly related to a damaged part included in the vehicle damage items is also recorded as a damaged part in the vehicle damage items, wherein, if the result of the identification is that it is not recorded, the damaged part is identified as a risk item; where the strong correlation means that if one vehicle part is damaged and thus becomes a damaged part, the probability that another vehicle part will also become a damaged part is higher than a predetermined threshold. [4] Method for risk control in the assessment of vehicle accident damage according to claim 1, characterized by that the previously created risk algorithm model a vehicle original equipment standard equipment information model, wherein the vehicle original equipment standard equipment information model is created through big data standardization of vehicle original equipment information, where the vehicle original equipment information includes at least one of the following items: vehicle original equipment data, part codes, quantities of parts per vehicle, relationships between assemblies and individual parts, indicative part prices and quantities of auxiliary materials. [5] Method for risk control in the assessment of vehicle accident damage according to claim 4, characterized by , that the calculation of the vehicle damage items in the vehicle accident damage assessment report based on the vehicle original standard equipment information model includes the following: Identifying original vehicle equipment information that corresponds to the vehicle, based on the vehicle information using the vehicle original equipment standard equipment information model, and comparing the information on damaged parts in the vehicle damage items with the original vehicle equipment information, whereby, if the result of the comparison between the two is inconsistent, the damaged parts that are inconsistent are identified as risk items. [6] Method for risk control in the assessment of vehicle accident damage according to claim 1, characterized by, that the previously created risk algorithm model includes a vehicle part standard repair process model, wherein the vehicle part standard repair process model is created through big data standardization of technical repair data and vehicle part repair process standards. [7] Method for risk control in the assessment of vehicle accident damage according to claim 6, characterized by , that the calculation of vehicle damage items in the vehicle accident damage assessment report using the vehicle part standard repair process model includes the following: Comparing repair information of damaged parts in the vehicle damage positions with the technical repair data and the repair process standards based on the vehicle information using the vehicle part standard repair process model, whereby, if the result of the comparison of the two is inconsistent, the damaged parts that are inconsistent are identified as risk positions. [8] Methods for risk control in the assessment of vehicle accident damage according to claims 1 to 7, characterized by , that it further includes the following: Select, after the user has indicated risks, whether or not a risk correction should be made based on the description of each indicated risk position, whereby, if the risk correction is selected, a risk identification and assessment is carried out again using the risk algorithm model. [9] Device for risk control in the assessment of vehicle accident damage, comprising: a recording unit for recording a report on the assessment of vehicle accident damage of a current accident of a vehicle, wherein the report on the assessment of vehicle accident damage contains vehicle information of the vehicle and vehicle damage items relating to the current accident of the vehicle; a risk identification and assessment unit for calculating the vehicle damage items in the vehicle accident damage assessment report using a pre-existing risk algorithm model in combination with the vehicle information, in order to identify risk items that pose a risk from the vehicle damage items, whereby the risk items are assessed as not belonging to the current vehicle accident or as inappropriate; and a risk processing unit for intercepting risk positions or user indications of risk positions, wherein The previously created risk algorithm model includes a vehicle collision damage level algorithm model, the vehicle collision damage level algorithm model being created in the following steps: Manually marking collision points and damage levels of historical vehicle accident cases using historical reports on the assessment of vehicle accident damage of various vehicle types and in combination with damage photos from historical vehicle accident damage assessments, and Learning historical reports on the assessment of vehicle accident damage of different vehicle types, as well as profiles of collision points and damage levels, using a computer in a machine-based manner based on manually marked collision points and damage levels of historical vehicle accident cases, in order to obtain correlations between the collision points and damage levels and the vehicle damage positions, and whereby The calculation of the vehicle damage items in the vehicle accident damage assessment report using the vehicle collision point damage level algorithm model includes the following: Calculating the damage level of the collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report and the collision point, calculated by the vehicle collision point algorithm model in the risk algorithm model, thereby determining a damage limit of the parts with respect to the collision point and the damage level, in order to assess whether damaged parts include a damaged part that lies outside the damage limit, whereby if the result of the assessment is that such a part is included, the damaged part that lies outside the damage limit is identified as a risk item. wherein the previously created risk algorithm model includes the vehicle collision point algorithm model, wherein the vehicle collision point algorithm model is created in the following steps: Virtual mapping of the vehicle into a corresponding diagram in a three-dimensional coordinate system according to the vehicle information and assignment of three-dimensional coordinates to vehicle parts according to the actual installation positions of the vehicle parts in the vehicle, Dividing the vehicle into different collision points and creating three-dimensional coordinate data for each of these collision points according to the vehicle information, thereby establishing relationships between the collision points and vehicle parts, where the calculation of the vehicle damage items in the report on the assessment of vehicle accident damage using the vehicle collision point algorithm model includes at least one of the following: Calculating a collision point of the current accident based on the vehicle damage items in the recorded current vehicle accident damage assessment report for the vehicle, and comparing the current vehicle accident damage assessment report for the vehicle with historical vehicle accident damage assessment reports for the vehicle to calculate whether the same collision point or damaged part is present, whereby, if the result of the calculation is that it is present, the same collision point or damaged part is identified as a risk item; Assess whether, if it is calculated that there are multiple collision points of the current accident of the vehicle, there is a collision point among the collision points that does not conform to the collision logic, whereby, if the result of the assessment is that such a collision point exists, the collision point that does not conform to the collision logic is identified as a risk position. [10] Electronic device, characterized by , that it one or more processors; a storage device for storing one or more programs includes wherein, when the one or more programs are executed by the one or more processors, this results in the one or more processors implementing a method according to any one of claims 1 to 8. [11] Computer-readable medium on which a computer program is stored, characterized bythat the program, when executed by a processor, implements a method according to any one of claims 1 to 8.

Images