Vehicle scrapping prediction method and device and vehicle management system

Abstract

The invention provides a vehicle scrapping prediction method and device and a vehicle management system, and relates to the technical field of computers.The prediction method comprises the steps that vehicle basic data of a to-be-detected vehicle are obtained, and the vehicle basic data comprise vehicle condition data, technical condition data, accident record data and maintenance records; obtaining a comprehensive evaluation index based on the technical condition data, the accident record data and the maintenance record; obtaining use environment data, and performing feature extraction on the use environment data and the vehicle condition data to obtain temporary feature data; inputting the temporary feature data, the comprehensive evaluation index and the vehicle condition data into a preset scrapping prediction model to obtain a scrapping prediction result; according to the method, the vehicle basic data and the use environment data are combined, and the prediction model is constructed by using the comprehensive evaluation index and the feature extraction technology, so that the scrapping risk of the vehicle can be predicted more accurately, and the scrapping prediction accuracy of the vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a vehicle scrapping prediction method, device and vehicle management system. Background Art

[0002] With the rapid development of economy, the number of motor vehicles is increasing. For vehicle manufacturers, understanding the needs of car buyers and accurately predicting the scrapped vehicles in the market are crucial to vehicle production planning.

[0003] In practical applications, with the diversification of vehicle types and the growth of data size, the scalability of traditional vehicle scrapping prediction methods has become particularly important. For example, vehicles of different brands, models or driving scenarios may have different scrapping characteristics and patterns.

[0004] However, traditional scrappage prediction methods are usually based on limited static data for analysis, such as vehicle age, mileage, etc., or only consider local specific vehicle characteristics as the basis for prediction, resulting in low accuracy of prediction results. Summary of the invention

[0005] The problem solved by the present invention is how to improve the accuracy of vehicle scrapping prediction.

[0006] In order to solve the above problems, the present invention provides a vehicle scrapping prediction method, comprising: Obtaining basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data, and repair and maintenance records; Obtaining a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; Acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data; The temporary characteristic data, the comprehensive evaluation index and the vehicle condition data are input into a preset scrapping prediction model to obtain a scrapping prediction result.

[0007] Optionally, the comprehensive evaluation index obtained based on the technical condition data, the accident record data and the repair and maintenance record includes: Preprocessing the technical condition data, the accident record data and the maintenance record respectively, and splicing the processed data to obtain temporary matrix data; Obtaining a preset indicator matrix, and obtaining a weighted standardized decision matrix based on the preset indicator matrix and the temporary matrix data; Based on preset positive and negative ideal solutions, the comprehensive evaluation index is determined according to the weighted standardized decision matrix.

[0008] Optionally, the preset positive and negative ideal solutions include a positive ideal solution and a negative ideal solution, and the step of determining a comprehensive evaluation index based on the preset positive and negative ideal solutions and the weighted standardized decision matrix includes: Respectively calculating the distance between the weighted normalized decision matrix and the positive ideal solution and the distance between the weighted normalized decision matrix and the negative ideal solution to obtain first distance data and second distance data; Obtaining the comprehensive evaluation index according to the first distance data and the second distance data through formula 1; Wherein, the formula 1 is: ; in, represents the first distance data between the vehicle to be tested X and the positive ideal solution P, represents the second distance data between the vehicle to be tested X and the negative ideal solution N, is the comprehensive evaluation index.

[0009] Optionally, the vehicle condition data includes vehicle age and mileage, the use environment data includes weather data, and the obtaining of the use environment data and performing feature extraction on the use environment data and the vehicle condition data to obtain temporary feature data includes: Obtaining the maintenance cycle of the vehicle to be tested, and determining the time distance to the next maintenance according to the vehicle age and the maintenance cycle; The driving mileage, the time distance and the weather data are integrated to obtain the temporary feature data.

[0010] Optionally, the usage environment data further includes road conditions, and the acquiring of the usage environment data and performing feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data further includes: The first characteristic data is obtained by calculating the vehicle age and the weather data, and the second characteristic data is obtained by calculating the mileage and the road condition; The first feature data and the second feature data are merged to obtain the temporary feature data.

[0011] Optionally, the process of constructing the preset scrap prediction model includes: Obtaining a labeled training data set, wherein the training data set includes historical feature information of historical vehicles and corresponding scrap status labels; An initial prediction model based on a random forest algorithm is constructed, and the initial prediction model is trained and tuned based on the training data set, and the adjusted initial prediction model is used as the preset scrap prediction model.

[0012] Optionally, the training and tuning the initial prediction model based on the training data set, and using the adjusted initial prediction model as the preset scrap prediction model, includes: Training the initial prediction model using the training data set to obtain a temporary prediction result; Perform loss calculation based on the temporary prediction result and the corresponding scrap status label to obtain a loss function output; The model parameters of the initial prediction model are adjusted according to the loss function output until the loss function input meets the preset conditions, and the initial prediction model after the parameter adjustment is used as the preset scrap prediction model.

[0013] The vehicle scrapping prediction method of the present invention has the following beneficial effects: The process of this vehicle scrapping prediction method is as follows: by obtaining the basic vehicle data and usage environment information of the vehicle to be tested, and based on the technical condition data, accident record data, and maintenance record information in the basic vehicle data, a comprehensive evaluation index is obtained. The overall status of the vehicle can be evaluated and analyzed through the comprehensive evaluation index, that is, the characteristics of the basic vehicle data and the usage environment data are comprehensively considered, and the scrapping risk of the vehicle can be evaluated more comprehensively and from multiple angles, thereby improving the accuracy and reliability of the prediction.

[0014] The environmental information and vehicle condition data are used for feature extraction. The environmental data used may include information such as the vehicle operating environment and driving road conditions. More feature data about the vehicle's operating status can be obtained through feature extraction. Finally, the temporary feature data, comprehensive evaluation indicators and vehicle condition data are input into the preset scrap prediction model. The scrap prediction model may be a model built based on machine learning or deep learning algorithms. The vehicle's scrap prediction results are obtained by training and predicting through input data. Based on comprehensive evaluation indicators and combined with the use of environmental data and vehicle condition data for feature extraction, the vehicle's operating status and environmental adaptability can be more accurately portrayed, providing the prediction model with richer feature information and improving the accuracy of the prediction results.

[0015] Therefore, the present invention combines vehicle basic data and usage environment data, and uses comprehensive evaluation indicators and feature extraction technology to build a prediction model, which can more accurately predict the vehicle's scrapping risk, improve the accuracy of vehicle scrapping prediction, and provide more effective management and maintenance reference, which has high practical value and social significance.

[0016] On the other hand, in order to solve the above problem, the present invention provides a vehicle scrapping prediction device, comprising: An acquisition unit, used to acquire basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data and repair and maintenance records; a processing unit, configured to obtain a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; The acquisition unit is further used to acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data; The processing unit is further used to input the temporary characteristic data, the comprehensive evaluation index and the vehicle condition data into a preset scrapping prediction model to obtain a scrapping prediction result.

[0017] The advantages of the vehicle scrapping prediction device described in the present invention and the vehicle scrapping prediction method described in the present invention over the prior art are the same, and will not be repeated here.

[0018] On the other hand, to solve the above problem, the present invention provides a vehicle management system, including a memory and a processor; The memory is used to store computer programs; The processor is used to implement the vehicle scrapping prediction method as described above when executing the computer program.

[0019] The advantages of the vehicle management system described in the present invention and the vehicle scrapping prediction method described in the present invention are the same as those of the prior art, which will not be described in detail here.

[0020] On the other hand, to solve the above problem, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the vehicle scrapping prediction method as described above is implemented.

[0021] The advantages of the computer-readable storage medium of the present invention and the vehicle scrapping prediction method over the prior art are the same and will not be described in detail herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic diagram of a process flow of a vehicle scrapping prediction method according to an embodiment of the present invention; Figure 2 A schematic diagram of a process of constructing a preset scrap prediction model according to an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a vehicle scrapping prediction device according to an embodiment of the present invention; Figure 4 The figure is a schematic diagram of the structure of a vehicle management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0024] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0025] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0026] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0027] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0028] like Figure 1 As shown, a vehicle scrapping prediction method provided by an embodiment of the present invention includes: S100, obtaining basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data, and repair and maintenance records.

[0029] Specifically, vehicle condition data may include the age and mileage of the vehicle. Technical condition data include the vehicle's engine operating status, brake system, suspension system and other technical performance. These data can be obtained through vehicle technical testing instruments, such as engine testing, brake force testing, etc. Accident record data: records whether the vehicle has ever had an accident, collision or damage. These data can be obtained through records from insurance companies, repair shops or transportation departments. Repair and maintenance records: records the maintenance of the vehicle, including regular maintenance, replacement of parts, repair history, etc. These data can be obtained through vehicle repair and maintenance record sheets or repair shop records.

[0030] By obtaining basic vehicle data, we can fully understand the overall condition and historical maintenance of the vehicle, providing an important basis for subsequent scrapping prediction and management decisions.

[0031] S200, obtaining a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record.

[0032] Specifically, when comprehensive evaluation indicators are obtained based on technical condition data, accident record data, and maintenance record data, a comprehensive score or indicator can be calculated based on the weight and importance of each data to comprehensively evaluate the overall condition and scrap risk of the vehicle. According to the importance and impact of the technical condition, accident record, and maintenance record, an evaluation indicator system is formulated. For example, the evaluation indicators, weights, and scoring criteria of each data can be used. First, the data is normalized to unify the data of different dimensions into the same dimensional range for subsequent calculations. According to the formulated evaluation indicator system and weights, the overall score or indicator of the vehicle is comprehensively calculated to reflect the overall condition and risk level of the vehicle.

[0033] Through comprehensive evaluation indicators, various information such as the vehicle's technical condition, accident record, and maintenance record can be taken into consideration to evaluate the vehicle's overall condition more comprehensively and objectively.

[0034] S300, obtaining usage environment data, and performing feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data.

[0035] Specifically, the use environment data includes information such as the geographical location of the vehicle, road conditions, climate conditions, and traffic conditions. These data can be obtained through GPS positioning systems, weather station data, traffic monitoring data, etc. The use environment data and vehicle condition data are subjected to feature extraction to extract features related to vehicle safety and reliability. Common features include the traffic density of the vehicle's geographical location, the impact of climate conditions on vehicle performance, and the impact of road conditions on vehicle wear.

[0036] The feature extraction results of the environmental data and vehicle condition data are combined to obtain temporary feature data for further analysis and prediction of the vehicle's operating conditions.

[0037] S400, inputting the temporary characteristic data, the comprehensive evaluation index and the vehicle condition data into a preset scrapping prediction model to obtain a scrapping prediction result.

[0038] Specifically, the temporary feature data, comprehensive evaluation indicators and vehicle condition data are integrated to ensure that the data format is consistent, missing values ​​are handled completely, and feature scaling or conversion is performed as needed. Select a model suitable for scrapping prediction, such as logistic regression, support vector machine, decision tree, etc. Use the prepared training data set to train the selected model so that it learns the relationship and pattern between the data. Among them, the scrapping prediction result obtained is usually a probability value, indicating the possibility of the vehicle being scrapped. For example: the model outputs a scrapping prediction result of 0.75, indicating that the possibility of the vehicle being scrapped is 75%.

[0039] By inputting temporary characteristic data, comprehensive evaluation indicators and vehicle condition data into the preset scrapping prediction model, it can help vehicle managers more accurately predict the future scrapping risk of vehicles and make maintenance and management preparations in advance. This can avoid the losses and impacts caused by the sudden scrapping of vehicles, improve the utilization rate and safety of vehicles, and reduce maintenance costs and management pressure. The prediction results can also provide important references for decision makers to help them formulate appropriate vehicle renewal plans and maintenance strategies.

[0040] In this embodiment, the vehicle basic data and use environment information of the vehicle to be tested are obtained, and the comprehensive evaluation index is obtained based on the technical condition data, accident record data, and maintenance record information in the vehicle basic data. The overall status of the vehicle can be evaluated and analyzed through the comprehensive evaluation index, that is, the characteristics of the vehicle basic data and the use environment data are comprehensively considered, and the scrap risk of the vehicle can be evaluated more comprehensively and from multiple angles, thereby improving the accuracy and reliability of the prediction.

[0041] The environmental information and vehicle condition data are used for feature extraction. The environmental data used may include information such as the vehicle operating environment and driving road conditions. More feature data about the vehicle's operating status can be obtained through feature extraction. Finally, the temporary feature data, comprehensive evaluation indicators and vehicle condition data are input into the preset scrap prediction model. The scrap prediction model may be a model built based on machine learning or deep learning algorithms. The vehicle's scrap prediction results are obtained by training and predicting through input data. Based on comprehensive evaluation indicators and combined with the use of environmental data and vehicle condition data for feature extraction, the vehicle's operating status and environmental adaptability can be more accurately portrayed, providing the prediction model with richer feature information and improving the accuracy of the prediction results.

[0042] Therefore, this embodiment combines vehicle basic data and usage environment data, and uses comprehensive evaluation indicators and feature extraction technology to build a prediction model, which can more accurately predict the vehicle's scrapping risk, improve the accuracy of vehicle scrapping prediction, and provide more effective management and maintenance reference, which has high practical value and social significance.

[0043] Optionally, the comprehensive evaluation index obtained based on the technical condition data, the accident record data and the repair and maintenance record includes: Preprocessing the technical condition data, the accident record data and the maintenance record respectively, and splicing the processed data to obtain temporary matrix data; Obtaining a preset indicator matrix, and obtaining a weighted standardized decision matrix based on the preset indicator matrix and the temporary matrix data; Based on preset positive and negative ideal solutions, the comprehensive evaluation index is determined according to the weighted standardized decision matrix.

[0044] Specifically, data cleaning, missing value processing, and outlier processing are performed on technical status data, accident record data, and maintenance records to ensure data quality. The pre-processed data is spliced ​​to form temporary matrix data for subsequent processing.

[0045] Obtain the preset indicator matrix to determine the indicators used for evaluation and organize them into a preset indicator matrix. Through the given indicators and temporary matrix data, they are weighted and standardized to obtain a weighted standardized decision matrix. Based on the set positive and negative ideal solutions, the weighted standardized decision matrix is ​​compared to obtain the final evaluation indicators.

[0046] For example, a decision matrix is ​​constructed through historical data (the vehicle's technical condition data, accident record data, and the repair and maintenance record), as shown in Table 1, the vehicle's decision matrix.

[0047] Table 1 Vehicle decision matrix

[0048] Normalize the values ​​of different indicators so that they are in the same range (usually 0 to 1). For example, the standardized technical condition data, accident record data, and maintenance records are shown in Table 2. The standardized decision data; Table 2 Standardized decision data

[0049] The preset positive ideal solution (the higher the score, the better) and the preset negative ideal solution (the lower the score, the better) are determined by the data in Table 2. As shown in Table 3, the preset positive ideal solution and the preset negative ideal solution.

[0050] Table 3 Preset positive ideal solution and preset negative ideal solution

[0051] By obtaining the preset indicator matrix, which can be the weight of each indicator, the normalized temporary matrix data is multiplied by the weight of each indicator. For example, the weight of technical condition data is 0.4, the weight of accident record data is 0.3, and the weight of maintenance record is 0.3, and a weighted standardized decision matrix is ​​obtained.

[0052] The distance between the weighted normalized decision matrix and the corresponding preset positive ideal solution and preset negative ideal solution is calculated to obtain a comprehensive evaluation index. The closer the comprehensive evaluation index is to 1, the better the quality of the vehicle is, and the closer it is to 0, the worse the quality is.

[0053] This process takes into account the vehicle's technical condition, accident record and maintenance status, and derives a comprehensive evaluation index from multiple perspectives. This helps vehicle managers to fully understand the vehicle's condition.

[0054] Optionally, the comprehensive evaluation index obtained based on the technical condition data, the accident record data and the repair and maintenance record further includes: The technical condition data, accident record data and maintenance record data are processed by fuzzy set theory to obtain comprehensive evaluation indicators, including: Specify fuzzy sets: First, the technical condition data, accident record data, and maintenance record data need to be converted into corresponding fuzzy sets to facilitate the processing of fuzzy information. For example, the evaluation of technical condition can be divided into "good", "average", and "poor", and each category has its membership function.

[0055] Determine fuzzy rules: Develop a set of fuzzy rules to describe the associations between different technical conditions, accident records, and maintenance situations. These rules can be based on expert knowledge or data analysis.

[0056] Fuzzy reasoning: Perform fuzzy reasoning on the vehicle's technical condition data, accident record data, and maintenance record data, and calculate the comprehensive evaluation value of each fuzzy set based on pre-determined fuzzy rules.

[0057] Defuzzification: Finally, the fuzzy comprehensive evaluation value needs to be defuzzified to convert the fuzzy result into a clear evaluation index. This can be achieved through common defuzzification methods such as the maximum membership method and the weighted average method.

[0058] For example, vehicle A has a technical condition score of "average" (membership degree 0.6), an accident record score of "small" (membership degree 0.8), and a repair and maintenance record score of "good" (membership degree 0.7). According to the fuzzy rules, we can get the following evaluation values: technical condition: 0.6 (average), accident record: 0.8 (small), repair and maintenance record: 0.7 (good).

[0059] Through fuzzy reasoning, we can calculate the comprehensive evaluation index as "medium", and then convert it into a clear evaluation value through defuzzification. For example, a score of 60 points (assuming the score is 0-100 points) indicates that the comprehensive evaluation of the car is at a medium level.

[0060] Optionally, the preset positive and negative ideal solutions include a positive ideal solution and a negative ideal solution, and the step of determining a comprehensive evaluation index based on the preset positive and negative ideal solutions and the weighted standardized decision matrix includes: Respectively calculating the distance between the weighted normalized decision matrix and the positive ideal solution and the distance between the weighted normalized decision matrix and the negative ideal solution to obtain first distance data and second distance data; Obtaining the comprehensive evaluation index according to the first distance data and the second distance data through formula 1; Wherein, the formula 1 is: ; in, represents the first distance data between the vehicle to be tested X and the positive ideal solution P, represents the second distance data between the vehicle to be tested X and the negative ideal solution N, is the comprehensive evaluation index.

[0061] Optionally, the vehicle condition data includes vehicle age and mileage, the use environment data includes weather data, and the obtaining of the use environment data and performing feature extraction on the use environment data and the vehicle condition data to obtain temporary feature data includes: Obtaining the maintenance cycle of the vehicle to be tested, and determining the time distance to the next maintenance according to the vehicle age and the maintenance cycle; The driving mileage, the time distance and the weather data are integrated to obtain the temporary feature data.

[0062] Specifically, collect the use environment data of the vehicle to be tested, such as weather data, for subsequent feature extraction and analysis. Perform feature extraction on the vehicle condition data, including vehicle age and mileage, as well as use environment data, and extract meaningful features for subsequent analysis. Use the age and maintenance cycle of the vehicle to be tested to determine the time distance to the next maintenance. It can also be determined through historical maintenance data and the vehicle manufacturer's recommended maintenance cycle.

[0063] The mileage, time distance and weather data are integrated to obtain temporary feature data. This may involve feature engineering techniques such as data standardization, feature scaling, feature selection, etc. to ensure the accuracy and usability of the data.

[0064] Optionally, the usage environment data further includes road conditions, and the acquiring of the usage environment data and performing feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data further includes: The first characteristic data is obtained by calculating the vehicle age and the weather data, and the second characteristic data is obtained by calculating the mileage and the road condition; The first feature data and the second feature data are merged to obtain the temporary feature data.

[0065] Specifically, the use environment data includes road conditions, which can be obtained through various sensors, road condition monitoring systems or third-party data sources. Road condition data may include information such as road slipperiness, road slope, and road traffic conditions. The first feature data is obtained by calculating based on the vehicle age and weather data. For example, the relationship between vehicle age and rainfall can be calculated to measure the impact of vehicle age on driving in rainy days.

[0066] The second characteristic data is calculated based on the mileage and road conditions. For example, the number of times the vehicle encounters slippery roads every 1,000 kilometers can be calculated to reflect the performance of the vehicle under different road conditions.

[0067] The first feature data and the second feature data are merged, and the comprehensive temporary feature data can be obtained by simple splicing, weighted average or feature crossover. For example, the feature value calculated by the vehicle age and weather data and the feature value calculated by the mileage and road conditions are spliced ​​to obtain the comprehensive temporary feature data.

[0068] By integrating temporary feature data with vehicle condition data and usage environment data, the actual operation of the vehicle is more comprehensively reflected. This comprehensive data information can provide more accurate and detailed feature descriptions, which helps to improve the accuracy and robustness of the prediction model. At the same time, combined with factors such as vehicle age, mileage, weather conditions and road conditions, we can better understand the vehicle's usage environment and operation. By comprehensively analyzing the relationship between different features, we can better discover the potential laws and trends of vehicle scrapping. The temporary feature data obtained contains richer and more diverse information, which can help build more complex and sophisticated prediction models. By utilizing information from multiple data sources, the accuracy and stability of vehicle scrapping predictions can be improved.

[0069] Alternatively, if Figure 2 As shown, the construction process of the preset scrap prediction model includes: Obtaining a labeled training data set, wherein the training data set includes historical feature information of historical vehicles and corresponding scrap status labels; An initial prediction model based on a random forest algorithm is constructed, and the initial prediction model is trained and tuned based on the training data set, and the adjusted initial prediction model is used as the preset scrap prediction model.

[0070] Optionally, the training and tuning the initial prediction model based on the training data set, and using the adjusted initial prediction model as the preset scrap prediction model, includes: Training the initial prediction model using the training data set to obtain a temporary prediction result; Perform loss calculation based on the temporary prediction result and the corresponding scrap status label to obtain a loss function output; The model parameters of the initial prediction model are adjusted according to the loss function output until the loss function input meets the preset conditions, and the initial prediction model after the parameter adjustment is used as the preset scrap prediction model.

[0071] Specifically, historical feature information of historical vehicles, such as vehicle age, mileage, maintenance records, etc., and corresponding scrap status labels (scrapped or not scrapped) are collected to form a training data set. For the collected training data set, an initial prediction model based on the random forest algorithm is established, which is used to predict the scrapping status of vehicles.

[0072] The initial prediction model is trained with the training data set to obtain a temporary prediction result. The temporary prediction result is compared with the actual scrap status label, and the output of the loss function is calculated, which can measure the error of the prediction result. According to the output of the loss function, the model parameters of the initial prediction model are adjusted, for example, by continuously adjusting the model parameters through optimization algorithms such as gradient descent until the loss function meets the preset conditions. When the loss function input meets the preset conditions, the initial prediction model after parameter adjustment is considered to be a trained preset scrap prediction model.

[0073] By continuously adjusting the model parameters according to the loss function, the accuracy and generalization ability of the model can be improved, thereby improving the accuracy of vehicle scrapping prediction. After establishing a preset scrapping prediction model, it can provide a more reliable decision-making basis and help managers formulate more reasonable vehicle management strategies and maintenance plans.

[0074] like Figure 3 As shown, a vehicle scrapping prediction device 300 provided by an embodiment of the present invention includes: The acquisition unit 310 is used to acquire basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data and repair and maintenance records; acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data.

[0075] The processing unit 320 is used to obtain a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; input the temporary feature data, the comprehensive evaluation index and the vehicle condition data into a preset scrapping prediction model to obtain a scrapping prediction result.

[0076] The vehicle scrapping prediction device of this embodiment is used to implement the vehicle scrapping prediction method as described above. Its advantages over the prior art are the same as the advantages of the above-mentioned vehicle scrapping prediction method over the prior art, and will not be repeated here.

[0077] Optionally, the processing unit 320 is specifically used to: pre-process the technical condition data, the accident record data and the maintenance record respectively, and splice the processed data to obtain temporary matrix data; obtain a preset indicator matrix, and obtain a weighted standardized decision matrix based on the preset indicator matrix and the temporary matrix data; based on preset positive and negative ideal solutions, determine the comprehensive evaluation index according to the weighted standardized decision matrix.

[0078] Optionally, the vehicle condition data includes vehicle age and mileage, the usage environment data includes weather data, and the acquisition unit 310 is specifically used to: obtain the maintenance cycle of the vehicle to be tested, and determine the time distance to the next maintenance based on the vehicle age and the maintenance cycle; merge the mileage, the time distance and the weather data to obtain the temporary feature data.

[0079] Optionally, the processing unit 320 is specifically used to obtain a labeled training data set, wherein the training data set includes historical feature information of historical vehicles and corresponding scrap status labels; construct an initial prediction model based on a random forest algorithm, and train and tune the initial prediction model based on the training data set, and use the adjusted initial prediction model as the preset scrap prediction model.

[0080] like Figure 4 As shown, a vehicle management system 400 provided by an embodiment of the present invention includes a memory 410 and a processor 420; the memory 410 is used to store a computer program; the processor 420 is used to implement the vehicle scrapping prediction method as described above when executing the computer program.

[0081] In other words, a vehicle management system 400 includes a memory 410 and a processor 420 coupled to the memory 410; the memory 410 is configured to store a computer program; and the processor 420 is configured to perform the following operations when executing the computer program: Obtaining basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data, and repair and maintenance records; Obtaining a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; Acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data; The temporary characteristic data, the comprehensive evaluation index and the vehicle condition data are input into a preset scrapping prediction model to obtain a scrapping prediction result.

[0082] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the vehicle scrapping prediction method as described above is implemented.

[0083] In other words, a non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following operations: Obtaining basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data, and repair and maintenance records; Obtaining a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; Acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data; The temporary characteristic data, the comprehensive evaluation index and the vehicle condition data are input into a preset scrapping prediction model to obtain a scrapping prediction result.

[0084] The vehicle management system 400 that can be used as the server or client of the present invention will now be described, which is an example of the hardware device that can be applied to all aspects of the present invention. The vehicle management system 400 is intended to represent the computer equipment of various forms of digital electronics, such as, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers and other suitable computers. The vehicle management system 400 can also represent various forms of mobile devices, such as, personal digital processing, cellular phones, smart phones, wearable devices and other similar computing devices. The components shown in this article, their connection and relationship, and their functions are only used as examples, and are not intended to limit the realization of the present invention described herein and / or required.

[0085] The vehicle management system 400 includes a computing unit that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The computing unit, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0086] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present invention. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0087] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A vehicle scrapping prediction method, characterized in that: include: Obtaining basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data, and repair and maintenance records; Obtaining a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; Acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data; The temporary characteristic data, the comprehensive evaluation index and the vehicle condition data are input into a preset scrapping prediction model to obtain a scrapping prediction result.

2. The vehicle scrapping prediction method according to claim 1, characterized in that: The comprehensive evaluation index obtained based on the technical condition data, the accident record data and the maintenance record includes: Preprocessing the technical condition data, the accident record data and the maintenance record respectively, and splicing the processed data to obtain temporary matrix data; Obtaining a preset indicator matrix, and obtaining a weighted standardized decision matrix based on the preset indicator matrix and the temporary matrix data; Based on preset positive and negative ideal solutions, the comprehensive evaluation index is determined according to the weighted standardized decision matrix.

3. The vehicle scrapping prediction method according to claim 2, characterized in that: The preset positive and negative ideal solutions include a positive ideal solution and a negative ideal solution, and the comprehensive evaluation index is determined based on the preset positive and negative ideal solutions and the weighted standardized decision matrix, including: Respectively calculating the distance between the weighted normalized decision matrix and the positive ideal solution and the distance between the weighted normalized decision matrix and the negative ideal solution to obtain first distance data and second distance data; Obtaining the comprehensive evaluation index according to the first distance data and the second distance data through formula 1; Wherein, the formula 1 is: ； in, represents the first distance data between the vehicle to be tested X and the positive ideal solution P, represents the second distance data between the vehicle to be tested X and the negative ideal solution N, is the comprehensive evaluation index.

4. The vehicle scrapping prediction method according to claim 1, characterized in that: The vehicle condition data includes vehicle age and mileage, the use environment data includes weather data, and the use environment data is obtained, and feature extraction is performed on the use environment data and the vehicle condition data to obtain temporary feature data, including: Obtaining the maintenance cycle of the vehicle to be tested, and determining the time distance to the next maintenance according to the vehicle age and the maintenance cycle; The driving mileage, the time distance and the weather data are integrated to obtain the temporary feature data.

5. The vehicle scrapping prediction method according to claim 4, characterized in that: The use environment data also includes road conditions, and the step of obtaining the use environment data and extracting features from the use environment data and the vehicle condition data to obtain temporary feature data also includes: The first characteristic data is obtained by calculating the vehicle age and the weather data, and the second characteristic data is obtained by calculating the mileage and the road condition; The first feature data and the second feature data are merged to obtain the temporary feature data.

6. The vehicle scrapping prediction method according to claim 1, characterized in that: The process of constructing the preset scrap prediction model includes: Obtaining a labeled training data set, wherein the training data set includes historical feature information of historical vehicles and corresponding scrap status labels; An initial prediction model based on a random forest algorithm is constructed, and the initial prediction model is trained and tuned based on the training data set, and the adjusted initial prediction model is used as the preset scrap prediction model.

7. The vehicle scrapping prediction method according to claim 6, characterized in that: The training and tuning of the initial prediction model based on the training data set, and using the adjusted initial prediction model as the preset scrap prediction model, includes: Training the initial prediction model using the training data set to obtain a temporary prediction result; Perform loss calculation based on the temporary prediction result and the corresponding scrap status label to obtain a loss function output; The model parameters of the initial prediction model are adjusted according to the loss function output until the loss function input meets the preset conditions, and the initial prediction model after the parameter adjustment is used as the preset scrap prediction model.

8. A vehicle scrapping prediction device, characterized in that: include: An acquisition unit, used to acquire basic vehicle data of the vehicle to be tested, wherein the basic vehicle data includes vehicle condition data, technical condition data, accident record data and repair and maintenance records; a processing unit, configured to obtain a comprehensive evaluation index based on the technical condition data, the accident record data and the repair and maintenance record; The acquisition unit is further used to acquire usage environment data, and perform feature extraction on the usage environment data and the vehicle condition data to obtain temporary feature data; The processing unit is further used to input the temporary characteristic data, the comprehensive evaluation index and the vehicle condition data into a preset scrapping prediction model to obtain a scrapping prediction result.

9. A vehicle management system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the vehicle scrapping prediction method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the vehicle scrapping prediction method according to any one of claims 1 to 7 is implemented.

Images