Method and system for predicting vehicle maintenance intention and electronic equipment

By predicting vehicle maintenance intentions and utilizing maintenance interval ranges and pre-trained models, the problem of car owners not being able to obtain vehicle maintenance information in a timely manner is solved, enabling timely reminders for vehicle maintenance and improving the user experience for car owners.

CN118917492APending Publication Date: 2024-11-08GREAT WALL MOTOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411212548.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Car owners are unable to obtain timely information about vehicle maintenance needs, causing inconvenience.

Method used

By obtaining the vehicle's maintenance interval range, and using a pre-trained maintenance intention prediction model, based on the vehicle's historical maintenance records and owner information, the system predicts whether the owner intends to perform maintenance, and outputs a prompt message when the owner does.

Benefits of technology

It enables timely reminders to car owners to perform maintenance within the vehicle maintenance interval, improving the user experience and avoiding the inconvenience of car owners manually recording mileage and duration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917492A_ABST
    Figure CN118917492A_ABST
Patent Text Reader

Abstract

The invention discloses a vehicle maintenance intention prediction method and system and electronic equipment, and the method comprises the steps: inputting the vehicle information of a vehicle into a model when the vehicle is in different time periods in a maintenance interval range, so as to predict whether a vehicle owner of the vehicle has a maintenance intention for the vehicle in the time period or not, so as to prompt the vehicle owner in time, and improve the maintenance efficiency of the vehicle. The vehicle owner does not need to know the information that the vehicle needs to be maintained in time, and the vehicle use experience of the vehicle owner is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of vehicle maintenance, and in particular to a method, system and electronic equipment for predicting vehicle maintenance intention. Background Art

[0002] During the use of the vehicle, the owner determines when the vehicle needs maintenance, usually based on the pre-determined mileage and driving time. This will result in the owner not being able to promptly learn that the vehicle needs maintenance, causing inconvenience to the owner. Summary of the invention

[0003] In view of this, the present application provides a method, system and electronic device for predicting vehicle maintenance intention to solve the problem in the prior art that the vehicle owner cannot be informed of the need for vehicle maintenance in a timely manner. The specific solution is as follows:

[0004] A method for predicting vehicle maintenance intention, comprising:

[0005] Obtaining a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle;

[0006] If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval range, inputting vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model;

[0007] Obtaining a prediction result output by the maintenance intention prediction model, wherein the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle during the first period of time;

[0008] If the prediction result indicates that the owner of the first vehicle intends to maintain the first vehicle during the first time period, maintenance reminder information is output.

[0009] Furthermore, if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval, the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event is input into a pre-trained maintenance intention prediction model to obtain a prediction result output by the maintenance intention prediction model, including:

[0010] If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is between the minimum value of the maintenance interval range and a first time, inputting vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model to obtain a first prediction result, where the first time is the sum of the minimum value of the maintenance interval range and a first specific time length;

[0011] If the first prediction result indicates that the owner of the first vehicle has no maintenance intention for the first vehicle, the first vehicle continues to travel;

[0012] If it is determined that the interval between the occurrence time of the last maintenance event and the current time is between the first time and the second time of the maintenance interval range, the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event is input into a pre-trained maintenance intention prediction model to obtain a second prediction result, where the second time is the sum of the first time and the first specific duration.

[0013] Furthermore, obtaining the maintenance interval range of the vehicle includes:

[0014] Obtaining historical maintenance record data of a plurality of vehicles, wherein the historical maintenance record data of each vehicle includes: an occurrence time corresponding to each maintenance event of each vehicle;

[0015] Maintenance interval ranges for the vehicles are determined based on historical maintenance record data for the plurality of vehicles.

[0016] Further, the determining of the vehicle maintenance interval range based on the historical maintenance record data of the plurality of vehicles includes:

[0017] Performing statistical analysis on the historical maintenance record data of the plurality of vehicles to determine the median of the interval between two adjacent maintenance events;

[0018] Determine the first quartile and the third quartile of the interval between the two adjacent maintenance events based on the median;

[0019] A maximum value and a minimum value of the maintenance interval range are obtained based on the first quartile and the third quartile.

[0020] Furthermore, the maintenance intention prediction model is trained, including:

[0021] Obtain vehicle information of each vehicle in the training sample set;

[0022] Determine a time period corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle in the training sample set in the maintenance interval range;

[0023] Model training is performed based on the vehicle information of each vehicle in the training sample set and the time period corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle in the maintenance interval range to obtain a trained maintenance intention prediction model.

[0024] Furthermore, inputting the vehicle information of the first vehicle into a pre-trained maintenance intention prediction model includes:

[0025] Processing the vehicle information of the first vehicle to obtain processed vehicle data, and inputting the processed vehicle data of the first vehicle into a maintenance intention prediction model that has been trained in advance;

[0026] The data processing includes at least one of the following: supplementation of missing values ​​in the vehicle information, binning of continuous data in the vehicle information, and unique hot encoding of categorical data in the vehicle information.

[0027] Furthermore, if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, outputting maintenance reminder information includes:

[0028] If the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determining a historical maintenance address in a historical maintenance record of the first vehicle;

[0029] The historical maintenance addresses are sorted in a specific order and then output.

[0030] Furthermore, if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, outputting maintenance reminder information includes:

[0031] If the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determining whether there is a start record of the first vehicle within a second specific time period before the current time;

[0032] If it is determined that the first vehicle has a start record within a second specific time period before the current time, maintenance reminder information is output.

[0033] A vehicle maintenance intention prediction system, comprising:

[0034] An obtaining unit, configured to obtain a maintenance interval range of the vehicle, wherein the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle;

[0035] a prediction unit, configured to input vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model when determining that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within a first time period within the maintenance interval; and obtain a prediction result output by the maintenance intention prediction model, wherein the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period;

[0036] The output unit is used to output maintenance reminder information when it is determined that the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle during the first time period.

[0037] An electronic device, comprising:

[0038] A processor is used to obtain a maintenance interval range of a vehicle, wherein the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in a first time period within the maintenance interval range, input the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a maintenance intention prediction model that has been trained in advance; obtain a prediction result output by the maintenance intention prediction model, wherein the prediction result indicates whether the owner of the first vehicle has a maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has a maintenance intention for the first vehicle in the first time period, output maintenance reminder information;

[0039] The memory is used to store the program for the processor to execute the above processing process.

[0040] It can be seen from the above technical scheme that the vehicle maintenance intention prediction method, system and electronic device disclosed in the present application obtain the vehicle maintenance interval range, which is the range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event is input into the maintenance intention prediction model that has been pre-trained to obtain the prediction result output by the maintenance intention prediction model, and the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle in the first time period, maintenance prompt information is output. In this scheme, when the vehicle is in different time periods within the maintenance interval range, the vehicle information of the vehicle is input into the model to predict whether the owner of the vehicle has maintenance intention for the vehicle in the time period, so as to prompt the owner in time, without the problem that the owner cannot obtain the information that the vehicle needs maintenance in time, thereby improving the vehicle use experience of the owner. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A flowchart of a method for predicting vehicle maintenance intention disclosed in an embodiment of the present application;

[0043] Figure 2 A flowchart of a method for predicting vehicle maintenance intention disclosed in an embodiment of the present application;

[0044] Figure 3 A flowchart of a method for predicting vehicle maintenance intention disclosed in an embodiment of the present application;

[0045] Figure 4 A flowchart of a method for predicting vehicle maintenance intention disclosed in an embodiment of the present application;

[0046] Figure 5 A schematic diagram of the structure of a vehicle maintenance intention prediction system disclosed in an embodiment of the present application;

[0047] Figure 6 A schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0050] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] During the use of the vehicle, the owner determines when the vehicle needs maintenance, usually based on the pre-determined mileage and driving time. This requires the owner to record the mileage and maintenance time of the last maintenance, and when the vehicle reaches the mileage and maintenance time for the next maintenance, determine which 4S store to go to for maintenance and find the address of the 4S store. This means that if the owner does not pay attention to the mileage or driving time, the vehicle cannot be maintained in time. That is, in the existing solution, the owner cannot be informed of the need for maintenance in time, which brings inconvenience to the owner.

[0052] Based on this, the present invention discloses a method for predicting vehicle maintenance intention. When it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval, the vehicle information of the first vehicle is input into a pre-trained maintenance intention prediction model, so that the maintenance intention prediction model can predict whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period within the maintenance interval, and when there is maintenance intention, maintenance reminder information can be output, so as to achieve the purpose of maintenance reminder for the owner of the first vehicle, without the need for the owner to remember the mileage and driving time himself, thereby improving the owner's experience.

[0053] The present application discloses a method for predicting vehicle maintenance intention, the flow chart of which is as follows: Figure 1 As shown, including:

[0054] Step S11, obtaining a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle;

[0055] Step S12: if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval, input the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into the pre-trained maintenance intention prediction model;

[0056] Step S13, obtaining a prediction result output by the maintenance intention prediction model, the prediction result indicating whether the owner of the first vehicle has maintenance intention for the first vehicle in the first period;

[0057] Step S14: if the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle in the first period, output maintenance reminder information.

[0058] The vehicle maintenance intention prediction method disclosed in this embodiment can be applied to the vehicle's on-board computer, and the vehicle's on-board computer predicts whether the owner of the current vehicle has the maintenance intention for the current vehicle. It can also be applied to the service end of the main engine manufacturer, or the service end of a 4S shop. The service end of the main engine manufacturer can predict whether the owners of all vehicles maintained by the main engine manufacturer have the maintenance intention for their vehicles, and the service end of the 4S shop can predict whether the owners of vehicles maintained at the 4S shop have the maintenance intention for their vehicles.

[0059] The vehicle maintenance interval range is determined based on multiple vehicles and is consistent with the maintenance records of multiple vehicles. Not every maintenance of each of the multiple vehicles complies with the maintenance interval range, but it complies with the intervals of a larger proportion of maintenance events in the multiple vehicles.

[0060] The vehicle's maintenance interval range refers to the range corresponding to the time interval between two adjacent maintenance events of the vehicle. The minimum value in the vehicle's maintenance interval range indicates the minimum time interval corresponding to the minimum value that the vehicle should have between two adjacent maintenance events. The maximum value in the vehicle's maintenance interval range indicates the maximum time interval corresponding to the maximum value that the vehicle should have between two adjacent maintenance events. For example: if the vehicle's maintenance interval range is 6 months to 9 months, then a vehicle should have its next maintenance at least 6 months after the current maintenance, and the next maintenance should be no later than 9 months after the current maintenance.

[0061] When it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the maintenance interval range, and when it is within the first time period within the maintenance interval range, the vehicle information of the first vehicle is input into a pre-trained maintenance intention prediction model to obtain a prediction result output by the maintenance intention prediction model.

[0062] The maintenance interval range can be divided into multiple different time periods. When the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in any of the time periods, the vehicle information of the first vehicle can be input into the maintenance intention prediction model to predict whether the owner of the first vehicle has the maintenance intention for the first vehicle during the time period, thereby achieving the purpose of accurate prediction.

[0063] The vehicle information of the first vehicle includes at least: vehicle attribute information, vehicle owner information, vehicle travel information and vehicle maintenance record, wherein the vehicle maintenance record records the occurrence time of the last maintenance record.

[0064] The vehicle attribute information may include: vehicle brand, vehicle purchase date, etc. Different vehicle brands may require different maintenance intervals for different parts of the vehicle. The vehicle purchase date can indicate whether the vehicle is new and the number of maintenance required.

[0065] The vehicle owner information may include: the age and gender of the vehicle owner. The maintenance frequency or vehicle maintenance time of vehicle owners of different ages or genders may be different. For example, for male vehicle owners aged 25-35, the vehicle maintenance frequency may be higher; for female vehicle owners aged 45-55, the vehicle maintenance frequency may be relatively lower.

[0066] The vehicle trip information may include: vehicle frame number vin, trip date, trip start time, trip end time, mileage at the start of the trip, mileage at the end of the trip, motor start time at the start of the trip, motor shutdown time at the end of the trip, etc. The vehicle trip information can have an impact on vehicle maintenance, such as: maintenance according to vehicle mileage, or maintenance according to vehicle driving time, etc.;

[0067] Vehicle maintenance records may include: vehicle frame number vin, maintenance order creation date, maintenance order creation time, maintenance amount, etc., among which the maintenance amount can indicate whether the maintenance is a deep maintenance, and it has a certain relationship with the time of the next maintenance. For example, when a vehicle was last maintained, it was determined to be a deep maintenance based on its maintenance amount. Then, the time when the owner performs the next maintenance on the vehicle will be extended from the time of the last maintenance. That is, after performing a deep maintenance, the interval between the deep maintenance and the next maintenance can be extended.

[0068] If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval, the vehicle information of the first vehicle is input into the maintenance intention prediction model. Since the vehicle information includes time information, such as: the purchase date of the vehicle, the time in the vehicle maintenance record, the time corresponding to the vehicle travel, etc., therefore, after the vehicle information of the first vehicle is input into the maintenance intention prediction model, it can be predicted whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period at the current time.

[0069] After obtaining the prediction result output by the maintenance intention prediction model, if the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle in the first time period, it is necessary to output maintenance reminder information to remind the owner that he has the intention to maintain the first vehicle in the current time period, and after determining that the first vehicle has a maintenance record in the current time period, restart the timer, so as to take the current maintenance record as the last maintenance event, and when it is determined that the interval between the occurrence time of the last maintenance event and the current time is within the maintenance interval range, re-prediction is performed to determine whether it is necessary to output the maintenance reminder information;

[0070] If the prediction result indicates that the owner of the first vehicle has no maintenance intention for the first vehicle in the first time period, the vehicle continues to travel. At this time, there is no need to output maintenance reminder information. Instead, as time goes by, when the next time period is reached, the prediction is continued to determine whether the owner of the first vehicle has maintenance intention for the first vehicle within the time period, so that when there is a maintenance intention, the reminder information can be output in time to prompt the owner to perform maintenance in time.

[0071] Specifically, if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is between the minimum value of the maintenance interval range and the first time, the vehicle information of the first vehicle is input into a pre-trained maintenance intention prediction model to obtain a first prediction result, where the first time is the sum of the minimum value of the maintenance interval range and the first specific time length; if the first prediction result indicates that the owner of the first vehicle has no maintenance intention for the first vehicle, the first vehicle continues to travel; if it is determined that the interval between the occurrence time of the last maintenance event and the current time is between the first time and the second time within the maintenance interval range, the vehicle information of the first vehicle is input into a pre-trained maintenance intention prediction model to obtain a second prediction result, where the second time is the sum of the first time and the first specific time length.

[0072] The maintenance interval range is divided into multiple different time periods, and the time length of each time period is the same. For example, if 30 days is a time period, then the period from the minimum value of the maintenance interval range to the minimum value of the maintenance interval range plus 30 days is a time period, the period from the minimum value of the maintenance interval range plus 30 days to the minimum value of the maintenance interval range plus 60 days is a time period, and the period from the minimum value of the maintenance interval range plus 60 days to the minimum value of the maintenance interval range plus 90 days is a time period, until the maximum value of the maintenance interval range is reached. If the maximum value and the minimum value of the maintenance interval range are 90 days, the maintenance interval range includes the three time periods in the above example. If the maximum value and the minimum value of the maintenance interval range are 70 days, the last time period in the maintenance interval range is from the minimum value of the maintenance interval range plus 60 days to the maximum value of the maintenance interval range.

[0073] After the last maintenance event of the first vehicle, as time goes by, the interval between the occurrence time of the last maintenance event of the first vehicle and the current time reaches the maintenance interval range, and the first time is the period of the minimum value of the maintenance interval range and the first time. The first event is the sum of the minimum value of the maintenance interval range and the first specific time length. If the above example is used for explanation, the first specific time length is 30 days. Of course, the first specific time length can also be other time lengths, which are not specifically limited here.

[0074] When the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is between the minimum value of the maintenance interval range and the first time, the vehicle information of the first vehicle is input into the maintenance intention prediction model to obtain a prediction result, so as to determine whether the owner of the first vehicle has the maintenance intention for the first vehicle within this period; if so, the maintenance prompt information is output, and after it is determined that the first vehicle has undergone this maintenance, the timing is restarted to calculate whether the duration of the end of this maintenance reaches the maintenance interval range; if there is no maintenance intention, when it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the next period within the maintenance interval range, the vehicle information of the first vehicle is again input into the maintenance intention model for prediction, and so on, until it is determined that the owner of the first vehicle has the maintenance intention for the first vehicle, or the time from the occurrence time of the last maintenance event of the first vehicle reaches the last period in the maintenance interval range.

[0075] The method for predicting vehicle maintenance intention disclosed in this embodiment obtains the vehicle maintenance interval range, which is the range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event is input into the maintenance intention prediction model that has been pre-trained, and the prediction result output by the maintenance intention prediction model is obtained, and the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle in the first time period, maintenance prompt information is output. In this solution, when the vehicle is in different time periods within the maintenance interval range, the vehicle information of the vehicle is input into the model to predict whether the owner has maintenance intention for the vehicle in different time periods within the maintenance interval range, thereby achieving the purpose of accurately predicting whether the owner has maintenance intention for the vehicle in different time periods; in addition, when it is predicted that the owner of the vehicle has maintenance intention for the vehicle in a certain time period, the owner is prompted in time to avoid the situation where the owner cannot obtain the information that the vehicle needs maintenance in time, thereby improving the vehicle use experience of the owner.

[0076] In another embodiment, after obtaining the maintenance interval range of the vehicle, as long as it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the maintenance interval range, the vehicle information of the first vehicle is input into the maintenance intention prediction model, and the maintenance intention prediction model predicts in which time period within the maintenance interval range the owner of the first vehicle has maintenance intention, and outputs maintenance reminder information when the time period is reached to prompt the owner to perform maintenance in time.

[0077] This embodiment discloses a method for predicting vehicle maintenance intention, and its flow chart is as follows: Figure 2 As shown, including:

[0078] Step S21, obtaining historical maintenance record data of a plurality of vehicles, wherein the historical maintenance record data of each vehicle includes: the occurrence time corresponding to each maintenance event of each vehicle;

[0079] Step S22, determining a vehicle maintenance interval range based on historical maintenance record data of a plurality of vehicles;

[0080] Step S23: if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first period within the maintenance interval, input the vehicle information of the first vehicle into a pre-trained maintenance intention prediction model;

[0081] Step S24, obtaining a prediction result output by the maintenance intention prediction model, the prediction result indicating whether the owner of the first vehicle has maintenance intention for the first vehicle in the first period;

[0082] Step S25: if the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle in the first period, output maintenance reminder information.

[0083] The historical maintenance record data of each vehicle at least includes: the occurrence time corresponding to each maintenance event of each vehicle. Based on the occurrence time corresponding to each maintenance event, the interval between two adjacent maintenance events can be determined.

[0084] Each vehicle will store its own historical maintenance record data. In addition, the vehicle manufacturer will also store the historical maintenance record data of the vehicles it maintains, and the 4S shop will also store the historical maintenance record data of the vehicles maintained in the shop. Therefore, both the vehicle manufacturer and the 4S shop can obtain the historical maintenance record data of multiple vehicles and analyze the historical maintenance record data of multiple vehicles to determine the maintenance interval range of the vehicles; in addition, while each vehicle stores its own historical maintenance record data, it can also obtain the historical maintenance record data of other vehicles with viewing authority from the vehicle manufacturer or 4S shop, so as to analyze the historical maintenance record data of multiple vehicles and determine the maintenance interval range of the vehicles.

[0085] Specifically, after obtaining the historical maintenance record data of multiple vehicles, the interval duration can be determined based on each two adjacent maintenance events of each vehicle, so that the interval durations of multiple adjacent maintenance events of the multiple vehicles can be obtained, and the maintenance interval range of the vehicles can be determined based on the obtained interval durations of multiple adjacent maintenance events of the multiple vehicles.

[0086] Outliers can be determined from the intervals between multiple adjacent maintenance events of multiple vehicles, that is, the intervals that obviously do not conform to most of the intervals. After removing these outliers, the maintenance interval range can be determined using the maximum and minimum values ​​of the remaining intervals; or, all the determined intervals are sorted on the duration axis to determine the maintenance interval range that conforms to most of the intervals.

[0087] It can also be more specific: statistically analyze the historical maintenance record data of multiple vehicles to determine the median of the interval between two adjacent maintenance events; determine the first quartile and the third quartile of the interval between two adjacent maintenance events based on the median; and obtain the maximum and minimum values ​​of the maintenance interval range based on the first quartile and the third quartile.

[0088] A statistical analysis is performed on the historical maintenance record data of multiple vehicles to determine the interval between each two adjacent maintenances of each vehicle, and all the interval durations obtained are sorted to determine the median, that is, the 50th percentile of all the interval durations arranged from small to large is determined, and this number is the median of all the interval durations. After determining the median, the first quartile (the 25th percentile of all the interval durations arranged from small to large) and the third quartile (the 75th percentile of all the interval durations arranged from small to large) are determined based on all the interval durations arranged from small to large. Then, the first quartile (the 25th percentile of all the interval durations arranged from small to large) and the third quartile (the 75th percentile of all the interval durations arranged from small to large) are used to calculate the maximum and minimum values ​​of the maintenance interval range, thereby determining the maintenance interval range.

[0089] The above method is used to determine the maintenance interval range, which can cover the interval between two adjacent maintenances for most vehicles and meets the maintenance requirements of most vehicles.

[0090] Among them, the maximum and minimum values ​​of the maintenance interval range are determined by using the first quartile and the third quartile, which can be: setting the first quartile to Q1 and the third quartile to Q3, the minimum value of the maintenance interval range is: Q3-(Q3-Q1)×1.5, and the maximum value of the maintenance interval range is: Q3+(Q3-Q1)×1.5.

[0091] After determining the maintenance interval range, it can be determined from a statistical perspective whether the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the maintenance interval range, that is, by means of a box plot, it is determined whether the value of the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is between the maximum value and the minimum value of the maintenance interval range in the box plot. If it is between the maximum value and the minimum value of the maintenance interval range in the box plot, it indicates that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the maintenance interval range; if it is not between the maximum value and the minimum value of the maintenance interval range in the box plot, it indicates that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is not within the maintenance interval range.

[0092] In addition, after determining the maintenance interval range, it is also possible to determine from a business perspective whether the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the maintenance interval range, that is, by comparing the value of the interval between the occurrence time of the last maintenance event of the first vehicle and the current time with the maximum and minimum values ​​of the maintenance interval range to determine whether it is within the maintenance interval range.

[0093] The method for predicting vehicle maintenance intention disclosed in this embodiment obtains historical maintenance record data of multiple vehicles, and the historical maintenance record data of each vehicle includes: the occurrence time corresponding to each maintenance event of each vehicle; determines the maintenance interval range of the vehicle based on the historical maintenance record data of multiple vehicles; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, inputs the vehicle information of the first vehicle into a maintenance intention prediction model that has been pre-trained; obtains the prediction result output by the maintenance intention prediction model, and the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle in the first time period, outputs maintenance prompt information. This solution determines the maintenance interval range of the vehicle through the historical maintenance record data of multiple vehicles, so that the maintenance interval range of the vehicle can meet the maintenance range of multiple vehicles, and the determined maintenance interval range can be applied to multiple vehicles to determine whether the conditions for inputting vehicle information into the maintenance intention prediction model are met, thereby avoiding the model prediction pressure caused by inputting vehicle information into the maintenance intention prediction model when the conditions are not met.

[0094] This embodiment discloses a method for predicting vehicle maintenance intention, and its flow chart is as follows: Figure 3 As shown, including:

[0095] Step S31, obtaining a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle;

[0096] Step S32: if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval, data processing is performed on the vehicle information of the first vehicle to obtain processed vehicle data, and the processed vehicle data of the first vehicle is input into the maintenance intention prediction model, wherein the data processing includes at least one of the following: supplementation of missing values ​​in the vehicle information, binning of continuous data in the vehicle information, and one-hot encoding of categorical data in the vehicle information;

[0097] Step S33, obtaining a prediction result output by the maintenance intention prediction model, the prediction result indicating whether the owner of the first vehicle has maintenance intention for the first vehicle in the first period;

[0098] Step S34: if the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle in the first period, output maintenance reminder information.

[0099] When it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within a first period within the maintenance interval, the vehicle information of the first vehicle is input into the maintenance intention prediction model.

[0100] Since the vehicle information of the first vehicle includes at least: vehicle attribute information, owner information, vehicle travel information and vehicle maintenance record, it includes not only the data that the model can recognize, but also some data that the model cannot recognize, or other data that needs to be preprocessed, such as: binning processing of continuous data, unique hot encoding of categorical data, supplementation of missing data, etc.

[0101] For example, the gender of the car owner in the car owner information is a categorical variable, which needs to be one-hot encoded to adjust the categorical data to numeric data; or the season in the vehicle trip information is also a categorical data, which needs to be one-hot encoded to adjust the categorical data to numeric data so that it can be applied in the model;

[0102] For example, the age in the car owner information is continuous data and needs to be binned, that is, different age groups are determined. When the car owner's age is obtained in the car owner information, the age group to which the car owner's age belongs is directly determined to make the model prediction more accurate, reduce the difficulty of model prediction, and avoid inputting each age into the model as a category.

[0103] For missing values, they need to be supplemented according to the pattern or type of the data where the missing values ​​are located.

[0104] After the vehicle information of the first vehicle is processed to obtain processed vehicle data, the vehicle data can be directly input into the maintenance intention prediction model so that the maintenance intention prediction model predicts whether the current owner has maintenance intention for the first vehicle based on the vehicle data.

[0105] The maintenance intention prediction model is obtained through model training in advance. Specifically, the training to obtain the maintenance intention prediction model may include:

[0106] Obtain vehicle information of each vehicle in a training sample set; determine a time period in a maintenance interval corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle in the training sample set; perform model training based on the vehicle information of each vehicle in the training sample set and the time period in the maintenance interval corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle, and obtain a trained maintenance intention prediction model.

[0107] The training sample set contains information of multiple vehicles, which is pre-acquired information used for model training. The training sample set includes vehicles with different vehicle attributes, vehicles with owner information of different genders or ages, vehicles with different vehicle trip information, and vehicles with different vehicle maintenance records.

[0108] Since the training sample set includes vehicle information of multiple vehicles, and the vehicle information of each vehicle includes vehicle maintenance records, the time period corresponding to the interval between two adjacent maintenance of each vehicle in the maintenance interval range is determined. For example, the interval between two adjacent maintenance of a certain vehicle is in the first time period in the maintenance interval range, that is, it is in the time period of the minimum value in the maintenance interval range and the first time. The first time is the sum of the minimum value of the maintenance interval range and the first specific time length. Therefore, at the first time period in the maintenance interval range, the label corresponding to the maintenance record of the vehicle is marked as 1. If there is no maintenance record in a certain time period, the label is marked as 0. The labels corresponding to different time periods of different vehicles and the vehicle information of multiple vehicles are used as training data for model training, and a prediction model that can predict which time period in the maintenance interval range a certain vehicle has maintenance intention can be obtained.

[0109] In addition, when conducting model training, after obtaining the vehicle information of multiple vehicles in the training sample set, it is also necessary to perform data processing on the vehicle information of multiple vehicles to obtain processed vehicle data, and use the processed vehicle data for model training to ensure the accuracy of the trained model.

[0110] The vehicle data processing includes at least one of the following: supplementing missing values ​​in the vehicle information, binning the continuous data in the vehicle information, and unique-hot encoding the categorical data in the vehicle information.

[0111] During model training, the deep forest network hyperparameters can be tuned through grid search to build a better maintenance intention prediction model. In addition, features can be screened according to the feature importance ranking, and features with low feature importance can be deleted to build the final maintenance intention prediction model.

[0112] After the model is trained using the above method to obtain a trained maintenance intention prediction model, if it is to be predicted whether the first vehicle has a maintenance intention in the first time period, the vehicle information of the first vehicle is input into the maintenance intention prediction model to obtain a prediction result of the maintenance intention prediction model. If the prediction result is 1, it indicates that the owner of the first vehicle has a maintenance intention for the first vehicle in the first time period. If the prediction result is 0, it indicates that the owner of the first vehicle has no maintenance intention for the first vehicle in the first time period, and it is necessary to continue the prediction when the interval duration reaches the next time period in the maintenance interval range.

[0113] The method for predicting vehicle maintenance intention disclosed in the present embodiment obtains a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, data processing is performed on the vehicle information of the first vehicle to obtain processed vehicle data, and the processed vehicle data of the first vehicle is input into a maintenance intention prediction model, where the data processing includes at least one of the following: supplementation of missing values ​​in the vehicle information, binning processing of continuous data in the vehicle information, and unique hot encoding of categorical data in the vehicle information; a prediction result output by the maintenance intention prediction model is obtained, and the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; wherein the vehicle information includes at least: vehicle attribute information, owner information, vehicle travel information, and vehicle maintenance record, and the first time period is any time period in which the maintenance interval range is divided into a plurality of different time periods; if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle in the first time period, maintenance prompt information is output. Before inputting the vehicle data into the maintenance intention prediction model, this solution first processes the vehicle data so as to bin the continuous data in the vehicle data, perform one-hot encoding on the categorical data, and supplement the missing values ​​to ensure that the data input into the maintenance intention prediction model meets the model processing requirements and can successfully implement model prediction.

[0114] This embodiment discloses a method for predicting vehicle maintenance intention, and its flow chart is as follows: Figure 4 As shown, including:

[0115] Step S41, obtaining a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle;

[0116] Step S42: if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval, input the vehicle information of the first vehicle into a pre-trained maintenance intention prediction model;

[0117] Step S43, obtaining a prediction result output by the maintenance intention prediction model, the prediction result indicating whether the owner of the first vehicle has maintenance intention for the first vehicle in the first period;

[0118] Step S44: if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determine a historical maintenance address in the historical maintenance record of the first vehicle;

[0119] Step S45: sort the historical maintenance addresses in a specific order and output them.

[0120] When it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current event is within the first time period within the maintenance interval, the vehicle information of the first vehicle is input into the maintenance intention prediction model to obtain a prediction result of the maintenance intention prediction model.

[0121] If the prediction result indicates that the owner of the first vehicle intends to maintain the first vehicle in the first period, maintenance reminder information needs to be output, and the maintenance reminder information can be generated based on a historical maintenance address in a historical maintenance record of the first vehicle.

[0122] Specifically, when it is determined that the prediction result indicates that the owner of the first vehicle has an intention to maintain the first vehicle in the first time period, the historical maintenance record of the first vehicle is obtained, and the maintenance address of each maintenance event in the historical maintenance record of the first vehicle is obtained, that is, which 4S stores or manufacturers that can provide maintenance have been visited in the historical maintenance record of the first vehicle.

[0123] The historical maintenance addresses in the historical maintenance record of the first vehicle are sorted in a specific order, and then the sorting result is output as maintenance reminder information.

[0124] Sorting in a specific order may be: in chronological order, for example, while determining the historical maintenance address in the historical maintenance record of the first vehicle, it is also necessary to determine the time when the first vehicle went to the historical maintenance address for maintenance, and then sorting in order from near to far in time, for example, if the last maintenance was at the first address and the second to last maintenance was at the second address, then when sorting, the first address is placed before the second address, so as to achieve the purpose of prompting the vehicle owner of the maintenance address he has recently visited;

[0125] In addition, the sorting in a specific order can also be: sorting by the number of visits, such as: while determining the historical maintenance addresses in the historical maintenance record of the first vehicle, recording the number of visits of the first vehicle to each maintenance address, and sorting in order from most to least, such as: the first vehicle went to the third maintenance address 3 times and went to the second maintenance address 5 times, then when sorting, the second maintenance address is ranked before the third maintenance address;

[0126] Secondly, the sorting can be performed in a specific order or according to the distance from the first vehicle's frequently parked address. For example, if the first vehicle is parked at the first address more times and for a longer period of time, then when determining the historical maintenance addresses, the distances between different historical maintenance addresses and the first address are determined respectively. When sorting, the maintenance addresses that are close to the first address are placed in front, and the maintenance addresses that are far from the first address are placed in the back.

[0127] Furthermore, if when determining the historical maintenance record of the first vehicle, it is determined that the first vehicle has no maintenance record within the third specific period of time before the current time, it indicates that the first vehicle has not been maintained at the current vehicle factory or 4S shop within the third specific period of time before the current time. At this time, an address that can provide maintenance services and is close to the address where the first vehicle is often parked can be directly recommended to the first vehicle as maintenance reminder information output. At the same time, activity information such as coupons or discount coupons related to the address that can provide maintenance services can also be pushed to the first vehicle.

[0128] In addition, when a prediction result is obtained by using the maintenance intention prediction model, and the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle in the first period, when prompt information needs to be output, it may specifically include:

[0129] If the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle, determine whether there is a startup record of the first vehicle within a second specific time before the current time; if it is determined that the first vehicle has a startup record within the second specific time before the current time, output maintenance reminder information.

[0130] That is, maintenance reminder information is output to the first vehicle only when it is determined that the first vehicle has a startup record within the second specific time period before the current time. If it has no startup record within the certain time period, there is no need to output maintenance reminder information to it.

[0131] Among them, determining whether the first vehicle has a startup record can be: determining whether the first vehicle outputs a vehicle communication tbox signal. If the first vehicle has outputted a tbox signal within a second specific time period, it indicates that the first vehicle has a startup record within the second specific time period. If the first vehicle has not outputted a tbox signal within the second specific time period, it indicates that the first vehicle has no startup record within the second specific time period.

[0132] The method for predicting vehicle maintenance intention disclosed in this embodiment obtains the vehicle maintenance interval range, which is the range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, the vehicle information of the first vehicle is input into a maintenance intention prediction model that has been pre-trained; wherein the vehicle information at least includes: vehicle attribute information, owner information, vehicle travel information and vehicle maintenance record, and the first time period is any time period in which the maintenance interval range is divided into multiple different time periods; obtain the prediction result output by the maintenance intention prediction model, and the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determine the historical maintenance address in the historical maintenance record of the first vehicle; and output the historical maintenance address after sorting it in a specific order. When it is determined that the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, the solution outputs maintenance reminder information according to the historical maintenance record of the first vehicle, so as to achieve the goal of not only reminding the owner that maintenance is required in the current time period, but also recommending the address for vehicle maintenance to the owner.

[0133] This embodiment discloses a vehicle maintenance intention prediction system, and its structural diagram is shown in FIG. Figure 5 As shown, including:

[0134] An acquisition unit 51 , a prediction unit 52 and an output unit 53 .

[0135] The obtaining unit 51 is used to obtain the maintenance interval range of the vehicle, and the maintenance interval range is the range corresponding to the time interval between two adjacent maintenance events of the vehicle;

[0136] The prediction unit 52 is used for inputting the vehicle information of the first vehicle into a maintenance intention prediction model that has been trained in advance when it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within a first period within the maintenance interval; obtaining a prediction result output by the maintenance intention prediction model, the prediction result indicating whether the owner of the first vehicle has maintenance intention for the first vehicle in the first period;

[0137] The output unit 53 is used to output maintenance reminder information when it is determined that the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle in the first time period.

[0138] Furthermore, the prediction unit is used to:

[0139] If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is between the minimum value of the maintenance interval range and the first time, the vehicle information of the first vehicle is input into a pre-trained maintenance intention prediction model to obtain a first prediction result, where the first time is the sum of the minimum value of the maintenance interval range and the first specific duration; if the first prediction result indicates that the owner of the first vehicle has no maintenance intention for the first vehicle, the first vehicle continues to travel; if it is determined that the interval between the occurrence time of the last maintenance event and the current time is between the first time and the second time in the maintenance interval range, the vehicle information of the first vehicle is input into a pre-trained maintenance intention prediction model to obtain a second prediction result, where the second time is the sum of the first time and the first specific duration.

[0140] Furthermore, the obtaining unit is used to:

[0141] Historical maintenance record data of a plurality of vehicles are obtained, wherein the historical maintenance record data of each vehicle includes: an occurrence time corresponding to each maintenance event of each vehicle; and a maintenance interval range of the vehicle is determined based on the historical maintenance record data of the plurality of vehicles.

[0142] Furthermore, the obtaining unit is used to:

[0143] Statistical analysis is performed on historical maintenance record data of multiple vehicles to determine the median of the time interval between two adjacent maintenance events; the first quartile and the third quartile of the time interval between two adjacent maintenance events are determined based on the median; and the maximum and minimum values ​​of the maintenance interval range are obtained based on the first quartile and the third quartile.

[0144] Furthermore, the vehicle maintenance intention prediction system disclosed in this embodiment may also include:

[0145] A training unit is used to obtain vehicle information of each vehicle in a training sample set; determine a time period in a maintenance interval corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle in the training sample set; perform model training based on the vehicle information of each vehicle in the training sample set and the time period in the maintenance interval corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle, to obtain a trained maintenance intention prediction model.

[0146] Furthermore, the prediction unit is used to:

[0147] The vehicle information of the first vehicle is processed to obtain processed vehicle data, and the processed vehicle data of the first vehicle is input into a pre-trained maintenance intention prediction model; wherein the data processing includes at least one of the following: supplementation of missing values ​​in the vehicle information, binning processing of continuous data in the vehicle information, and unique hot encoding of categorical data in the vehicle information.

[0148] Furthermore, the output unit is used to:

[0149] If the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determine the historical maintenance addresses in the historical maintenance record of the first vehicle; and output the historical maintenance addresses after sorting them in a specific order.

[0150] Furthermore, the output unit is used to:

[0151] If the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle, determine whether there is a startup record of the first vehicle within a second specific time before the current time; if it is determined that the first vehicle has a startup record within the second specific time before the current time, output maintenance reminder information.

[0152] The vehicle maintenance intention prediction system disclosed in this embodiment is implemented based on the vehicle maintenance intention prediction method disclosed in the above embodiments, which will not be described in detail here.

[0153] The vehicle maintenance intention prediction system disclosed in this embodiment obtains the vehicle maintenance interval range, which is the range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, the vehicle information of the first vehicle is input into the maintenance intention prediction model that has been pre-trained to obtain the prediction result output by the maintenance intention prediction model, and the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle in the first time period, maintenance reminder information is output. This solution inputs the vehicle information of the vehicle into the model when the vehicle is in different time periods within the maintenance interval range, so as to predict whether the owner of the vehicle has maintenance intention for the vehicle in the time period, so as to promptly remind the owner, without the owner being unable to obtain the information that the vehicle needs maintenance in time, thereby improving the owner's vehicle use experience.

[0154] This embodiment discloses an electronic device, and its structural diagram is as follows: Figure 6 As shown, including:

[0155] Processor 61 and memory 62.

[0156] The processor 61 is used to obtain a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, the vehicle information of the first vehicle is input into a maintenance intention prediction model that has been pre-trained; a prediction result output by the maintenance intention prediction model is obtained, and the prediction result indicates whether the owner of the first vehicle has a maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has a maintenance intention for the first vehicle in the first time period, maintenance prompt information is output;

[0157] The memory 62 is used to store the program for the processor to execute the above-mentioned processing process.

[0158] The electronic device disclosed in this embodiment is implemented based on the vehicle maintenance intention prediction method disclosed in the above embodiments, which will not be described in detail here.

[0159] The electronic device disclosed in this embodiment obtains the vehicle's maintenance interval range, which is the range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is in the first time period within the maintenance interval range, the vehicle information of the first vehicle is input into the maintenance intention prediction model that has been pre-trained to obtain the prediction result output by the maintenance intention prediction model, and the prediction result indicates whether the owner of the first vehicle has the maintenance intention for the first vehicle in the first time period; if the prediction result indicates that the owner of the first vehicle has the maintenance intention for the first vehicle in the first time period, maintenance reminder information is output. This solution inputs the vehicle information of the vehicle into the model when the vehicle is in different time periods within the maintenance interval range, so as to predict whether the owner of the vehicle has the maintenance intention for the vehicle in the time period, so as to promptly remind the owner, without the owner being unable to obtain the information that the vehicle needs maintenance in a timely manner, thereby improving the owner's vehicle use experience.

[0160] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0161] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0162] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0163] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A method for predicting vehicle maintenance intention, characterized in that: include: Obtaining a maintenance interval range of the vehicle, where the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle; If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval range, inputting vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model; Obtaining a prediction result output by the maintenance intention prediction model, wherein the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle during the first period of time; If the prediction result indicates that the owner of the first vehicle intends to maintain the first vehicle during the first time period, maintenance reminder information is output.

2. The method according to claim 1, characterized in that If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within the first time period within the maintenance interval range, inputting vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model to obtain a prediction result output by the maintenance intention prediction model, including: If it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is between the minimum value of the maintenance interval range and a first time, inputting vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model to obtain a first prediction result, where the first time is the sum of the minimum value of the maintenance interval range and a first specific time length; If the first prediction result indicates that the owner of the first vehicle has no maintenance intention for the first vehicle, the first vehicle continues to travel; If it is determined that the interval between the occurrence time of the last maintenance event and the current time is between the first time and the second time of the maintenance interval range, the vehicle information of the first vehicle including at least the occurrence time of the last maintenance event is input into a pre-trained maintenance intention prediction model to obtain a second prediction result, where the second time is the sum of the first time and the first specific duration.

3. The method according to claim 1, characterized in that The obtaining of the vehicle maintenance interval range includes: Obtaining historical maintenance record data of a plurality of vehicles, wherein the historical maintenance record data of each vehicle includes: an occurrence time corresponding to each maintenance event of each vehicle; Maintenance interval ranges for the vehicles are determined based on historical maintenance record data for the plurality of vehicles.

4. The method according to claim 3, characterized in that The determining of the vehicle maintenance interval range based on the historical maintenance record data of the plurality of vehicles comprises: Performing statistical analysis on the historical maintenance record data of the plurality of vehicles to determine the median of the interval between two adjacent maintenance events; Determine the first quartile and the third quartile of the interval between the two adjacent maintenance events based on the median; A maximum value and a minimum value of the maintenance interval range are obtained based on the first quartile and the third quartile.

5. The method according to claim 1, characterized in that The maintenance intention prediction model is trained, including: Obtain vehicle information of each vehicle in the training sample set; Determine a time period corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle in the training sample set in the maintenance interval range; Model training is performed based on the vehicle information of each vehicle in the training sample set and the time period corresponding to the occurrence time of each maintenance event in the vehicle information of each vehicle in the maintenance interval range to obtain a trained maintenance intention prediction model.

6. The method according to claim 1, characterized in that The step of inputting the vehicle information of the first vehicle into a pre-trained maintenance intention prediction model includes: Processing the vehicle information of the first vehicle to obtain processed vehicle data, and inputting the processed vehicle data of the first vehicle into a maintenance intention prediction model that has been trained in advance; The data processing includes at least one of the following: supplementation of missing values ​​in the vehicle information, binning of continuous data in the vehicle information, and unique hot encoding of categorical data in the vehicle information.

7. The method according to claim 1, characterized in that If the prediction result indicates that the owner of the first vehicle intends to maintain the first vehicle, outputting maintenance reminder information includes: If the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determining a historical maintenance address in a historical maintenance record of the first vehicle; The historical maintenance addresses are sorted in a specific order and then output.

8. The method according to claim 1, characterized in that If the prediction result indicates that the owner of the first vehicle intends to maintain the first vehicle, outputting maintenance reminder information includes: If the prediction result indicates that the owner of the first vehicle has maintenance intention for the first vehicle, determining whether there is a start record of the first vehicle within a second specific time period before the current time; If it is determined that the first vehicle has a start record within a second specific time period before the current time, maintenance reminder information is output.

9. A vehicle maintenance intention prediction system, characterized in that: include: An obtaining unit, configured to obtain a maintenance interval range of the vehicle, wherein the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle; a prediction unit, configured to input vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model when determining that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within a first time period within the maintenance interval; and obtain a prediction result output by the maintenance intention prediction model, wherein the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; The output unit is used to output maintenance reminder information when it is determined that the prediction result indicates that the owner of the first vehicle has the intention to maintain the first vehicle during the first time period.

10. An electronic device, characterized in that: include: A processor is configured to obtain a maintenance interval range of a vehicle, wherein the maintenance interval range is a range corresponding to the time interval between two adjacent maintenance events of the vehicle; if it is determined that the interval between the occurrence time of the last maintenance event of the first vehicle and the current time is within a first time period within the maintenance interval range, input vehicle information of the first vehicle including at least the occurrence time of the last maintenance event into a pre-trained maintenance intention prediction model; obtain a prediction result output by the maintenance intention prediction model, wherein the prediction result indicates whether the owner of the first vehicle has maintenance intention for the first vehicle in the first time period; If the prediction result indicates that the owner of the first vehicle intends to maintain the first vehicle during the first period, outputting maintenance reminder information; The memory is used to store the program for the processor to execute the above processing process.