Automobile aftermarket intelligent management platform based on big data

Abstract

The invention discloses an automobile aftermarket intelligent management platform based on big data, and relates to the field of automobile data management, and the platform comprises the following modules: a data collection module which is used for collecting vehicle maintenance event data of a plurality of regions in real time; the space-time grid division module is used for dividing the to-be-analyzed area into a plurality of space grids according to a preset granularity and continuously dividing time into a plurality of time windows with fixed lengths; the data preprocessing module is used for mapping the vehicle maintenance event data into a corresponding space grid and a time window to generate a space-time event matrix; the maintenance type propagation modeling module is used for analyzing the space-time event matrix so as to simulate propagation trends of different maintenance types among the space grids along with time; and the maintenance activeness prediction module is used for predicting the maintenance activeness of the maintenance type corresponding to each space grid at the future moment. According to the invention, the problems of poor maintenance service prediction accuracy and low resource scheduling efficiency are solved.

Description

Technical Field

[0001] The present invention relates to the field of automobile data management, and more specifically, to an automobile aftermarket intelligent management platform based on big data. Background Art

[0002] The automotive aftermarket mainly includes services such as automobile repair, maintenance, and parts supply, and is an important part of the automotive industry chain. With the rapid growth of automobile ownership, the demand for repair services has increased significantly. However, traditional automotive aftermarket management relies mostly on manual statistics and experience-based decision-making, which makes it difficult to efficiently control the real-time changes in demand for repair services. It is prone to problems such as uneven distribution of repair resources and delayed services, especially in large and medium-sized cities.

[0003] In response to the above problems, existing technologies attempt to assist in predicting maintenance needs through data statistical analysis, such as conventional time series analysis, hot spot area identification and other methods, in order to achieve resource optimization scheduling. However, such methods are usually limited to static or simple statistical models, ignoring the propagation effect of maintenance activities in space and the dynamic change trend in time, making it difficult to accurately predict the rapid changes in maintenance needs in local areas. In addition, the currently widely used prediction technology also lacks in-depth research on the propagation laws between multiple maintenance types, resulting in the difficulty of existing technologies to accurately respond to maintenance needs in advance, and maintenance resources cannot be finely scheduled.

[0004] Therefore, the existing technology urgently needs a more efficient and accurate maintenance demand prediction solution to solve the problems of inaccurate maintenance service prediction and low resource allocation efficiency in the automotive aftermarket. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent management platform for the automotive aftermarket based on big data to solve the problems mentioned in the background technology.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: An intelligent management platform for the automotive aftermarket based on big data, including: The data collection module is used to collect vehicle maintenance event data in multiple areas in real time. Each maintenance event data includes the occurrence time, geographic location information and maintenance type; The space-time grid division module is used to divide the area to be analyzed into multiple space grids according to the preset granularity, and to divide the time continuously into multiple time windows of fixed length; A data preprocessing module, used for mapping the vehicle maintenance event data into corresponding spatial grids and time windows to generate a spatiotemporal event matrix; A maintenance type propagation modeling module is used to analyze the spatiotemporal event matrix, build and train a spatiotemporal graph convolutional network model to simulate the propagation trend of different maintenance types between spatial grids over time; The maintenance activity prediction module is used to predict the maintenance activity of the corresponding maintenance type of each spatial grid at future moments based on the trained spatiotemporal graph convolutional network model.

[0007] In some embodiments, the data collected by the data collection module specifically includes at least the latitude and longitude coordinates corresponding to the maintenance event, a timestamp, and a classification label of the maintenance type.

[0008] In some embodiments, the spatiotemporal grid division module divides the urban area into regular square spatial grids according to the geographic area boundaries of the city map and preset grid division rules, and each spatial grid corresponds to a unique spatial identifier.

[0009] In some embodiments, the area of ​​a single grid of the spatial grid division rule is between 250 meters×250 meters and 1 kilometer×1 kilometer.

[0010] In some embodiments, the data preprocessing module maps the maintenance event to the corresponding spatial grid and time window based on the latitude and longitude coordinates and timestamp of each maintenance event data, and accumulates and counts the event frequency of the maintenance type in each grid and each time window to generate a spatiotemporal event matrix with multi-dimensional maintenance type statistical values.

[0011] In some embodiments, the maintenance type propagation modeling module establishes a spatiotemporal graph structure, uses spatial grids as graph nodes, and sets edges between adjacent grids for connection; the maintenance type statistics of each node in different time windows are used as node attributes to construct a spatiotemporal graph network structure.

[0012] In some embodiments, the spatiotemporal graph convolutional network model used by the maintenance type propagation modeling module includes a spatial graph convolutional layer and a temporal convolutional layer; The spatial graph convolution layer is used to learn the spatial correlation between nodes in different spatial grids, and the temporal convolution layer is used to capture the trend of each node changing over time.

[0013] In some embodiments, the maintenance activity prediction module is based on a trained spatiotemporal graph convolutional network model and predicts the maintenance activity values ​​of various maintenance types in each spatial grid in the next time window or multiple consecutive time windows in the future according to the maintenance activity data at the current moment and historical moments.

[0014] In some embodiments, the maintenance activity value predicted by the maintenance activity prediction module is used to generate a maintenance type heat map, and the heat map is used to intuitively display the activity level of maintenance types in different spatial grids at future moments.

[0015] In some embodiments, the platform also includes a maintenance resource scheduling module for pre-allocating maintenance personnel, maintenance equipment, and spare parts inventory at automobile maintenance outlets based on the prediction results of the maintenance activity prediction module to cope with upcoming high-incidence maintenance areas and maintenance peak periods.

[0016] The present invention provides an intelligent management platform for the automotive aftermarket based on big data. It utilizes real-time collected vehicle maintenance event data, combined with spatiotemporal grid division and graph convolution deep learning models, to achieve accurate prediction of maintenance service needs, effectively solving the problems of insufficient prediction accuracy and lagging resource allocation in traditional solutions.

[0017] Specifically, the platform of the present invention obtains the occurrence time, geographical location and maintenance type data of maintenance events in real time, constructs a high-precision spatiotemporal event matrix, and effectively depicts the detailed distribution of maintenance events in different regions and times; on this basis, the spatiotemporal graph convolutional network model is introduced to successfully capture the propagation effects and time evolution trends of different maintenance types between spatial grids, thereby significantly improving the accuracy and reliability of maintenance activity prediction.

[0018] Furthermore, the platform of the present invention enhances the spatial resolution of the prediction results through the reasonable division of spatial grids and detailed maintenance event frequency statistics; at the same time, the prediction module supports multi-step prediction and independent modeling of multiple types of maintenance events, and intuitively displays the prediction results with the help of heat maps, so that managers can grasp the dynamic change trend of maintenance hotspot areas at a glance. In addition, the platform of the present invention also has the function of intelligent scheduling of maintenance resources in advance based on the prediction results, thereby effectively improving the utilization efficiency of maintenance personnel, maintenance equipment and spare parts inventory, reducing maintenance response delays, and significantly improving the overall maintenance service quality and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the module of the present invention; Figure 2 It is a flow chart of data preprocessing and spatiotemporal event matrix generation of the present invention; Figure 3 It is a flow chart of model construction and training of the present invention; Figure 4 It is a maintenance activity prediction and heat map generation flow chart of the present invention. DETAILED DESCRIPTION

[0020] The specific implementation of the present invention will be described below in conjunction with the accompanying drawings.

[0021] The present invention provides an intelligent management platform for the automotive aftermarket based on big data. The design idea of ​​the platform of the present invention is to build an adaptive modeling structure with the help of the spatial and temporal distribution characteristics of vehicle maintenance big data, so as to predict the evolution trend of maintenance behavior and provide data support for subsequent resource scheduling.

[0022] As shown in FIG1 , the platform of the present invention includes functional components such as a data acquisition module, a spatiotemporal grid division module, a data preprocessing module, a maintenance type propagation modeling module, and a maintenance activity prediction module.

[0023] The data collection module is responsible for obtaining maintenance event data in real time from vehicle maintenance outlets or related information systems in multiple regions. The collected data includes but is not limited to the time of occurrence of each maintenance event, the latitude and longitude coordinates of the maintenance location, and the corresponding maintenance type label. The classification of maintenance types can be divided according to industry practices, such as engine failure, brake system maintenance, body repair, electrical system maintenance, etc. Through the continuous operation of this module, a large-scale, continuously updated maintenance event database can be constructed.

[0024] In order to facilitate the modeling and analysis of the regional distribution and time evolution of maintenance events, it is necessary to standardize the time and space dimensions in the original data. In this embodiment, the spatiotemporal grid division module divides the urban area into spatial grids according to the actual boundaries of the city map according to preset rules. The shape of the spatial grid is usually square to simplify the subsequent calculation and processing. Each grid has a unique identifier for subsequent positioning and reference in the model.

[0025] More specifically, the appropriate grid granularity can be set according to the city size, road density, and spatial distribution density of maintenance events. In practice, the area of ​​a single grid is usually set between 250 meters × 250 meters and 1 km × 1 km. Smaller grids help capture finer-grained maintenance event distribution characteristics and are suitable for the core areas of large cities with high event density and rapid spatial changes. In suburban areas or areas where maintenance events are sparsely distributed, larger grids can be used to reduce computing resource consumption.

[0026] In terms of the processing of the time dimension, the entire observation period can be continuously divided into multiple time windows of fixed length. The length of the time window can be set according to the frequency of maintenance events, for example, 30 minutes, 1 hour or 1 day. Each maintenance event is classified into the corresponding time window according to its timestamp, forming a data structure that can be used for time series modeling.

[0027] After the space and time division is completed, each maintenance event data needs to be accurately mapped to the corresponding space grid and time window. As shown in Figure 2, this process is completed by the data preprocessing module. The specific method is: First, read the longitude and latitude information of each maintenance event, and determine the spatial grid number where the event is located by combining the geographic boundary information of the spatial grid; Secondly, read the timestamp information of the event, and perform a time window operation on it to determine its corresponding time window number.

[0028] Subsequently, in three-dimensional space, a statistical matrix is ​​established with “spatial grid × time window × maintenance type” as the coordinate axes.

[0029] Every time a maintenance event occurs, the system increments the frequency of events of that type in the corresponding cell. By processing all historical maintenance events, a spatiotemporal event matrix with maintenance type dimensions can be obtained. This matrix not only reflects the maintenance activity of a certain area in a certain period of time, but also reveals the evolution trend of different types of maintenance events, providing rich spatiotemporal feature information for subsequent modeling.

[0030] The corresponding formula can be expressed as: ; in, represents the maintenance event frequency on the spatial grid number s, time window number t, and maintenance type number k; N is the total number of maintenance events; It is The spatial grid number to which the dimension repair event belongs. It is The time window number to which the maintenance event belongs. and The maintenance type number to which the maintenance event belongs; is an indicator function that takes the value 1 when the condition in the brackets is met, otherwise it takes the value 0.

[0031] In order to further explore the propagation law of maintenance behavior in urban space, as shown in Figure 3, the present invention establishes a data representation method based on graph structure, abstracting each spatial grid as a node in the graph. The connection relationship between nodes is determined based on spatial adjacency, that is, if two spatial grids are geographically adjacent, a connecting edge is set for them in the graph structure. This mapping method retains the local adjacency relationship in the urban spatial structure, so that the subsequent graph neural network can learn the influence of maintenance behavior between adjacent areas. The attributes of each node in different time windows are composed of the corresponding maintenance type statistics, that is, the frequency of various maintenance events of the node in the time window. Through this mapping method, the originally isolated spatiotemporal event matrix can be converted into a dynamic graph network with structural information, providing structured input for the propagation modeling of maintenance behavior.

[0032] In practical applications, this graph structure not only supports static modeling, but can also be dynamically updated over time. As new maintenance events are continuously collected, node attributes will be continuously updated, and the graph structure can also adjust edge weights or partially reconstruct according to the transfer of maintenance event hotspots. The establishment of this graph network provides a natural structural foundation for subsequent deep learning models based on spatiotemporal graph convolution, enabling the model to simultaneously capture the mutual influence between spatial neighborhoods and the temporal evolution trend, which helps to achieve high-precision maintenance activity prediction and behavior evolution analysis.

[0033] More specifically, by combining the advantages of graph neural networks and time series modeling, the synergy of spatial graph convolutional layers and temporal convolutional layers can achieve joint learning of maintenance behaviors in both spatial and temporal dimensions.

[0034] The spatial graph convolution layer is mainly used to capture the correlation between adjacent spatial grid nodes. In an urban environment, maintenance behaviors between different areas are not completely independent. High-frequency maintenance events in a certain area often have an impact on surrounding areas. For example, when brake system maintenance occurs frequently along a traffic artery, it may mean that the road conditions are abnormal, or that a certain type of vehicle has a structural fault in a specific area. This pattern will show a certain degree of spatial continuity. Through graph convolution operations, the information of neighboring nodes can be effectively fused, and the cross-regional maintenance behavior propagation path can be learned.

[0035] The temporal convolution layer is used to extract the changing trend of each spatial node in the time dimension. The activity of maintenance events often has obvious periodicity and suddenness, which may change with the day-night rhythm, weekdays and weekends, and may also be affected by external factors such as weather, holidays, and promotions. The temporal convolution layer performs convolution operations on the time series of each node in units of time windows, which helps the model identify dynamic features such as rising trends, sudden increases, or continuous low frequencies.

[0036] After the model training is completed, it can be applied to the task of predicting maintenance activity.

[0037] As shown in Figure 4, the platform uses the model to input maintenance event statistics within the historical time window, and combines the current state to predict the activity values ​​of various maintenance types in different spatial grids within a future time window or multiple consecutive time windows. These predicted values ​​essentially reflect the possible density of maintenance activities of different types at different times and locations, and have strong spatiotemporal orientation and business guidance significance.

[0038] In some embodiments, in order to present the prediction results more intuitively to management personnel and operation and maintenance teams, the predicted maintenance activity values ​​can be converted into maintenance type heat maps. In the heat map, different colors or brightness represent different activity levels, allowing users to identify high-risk areas, potential maintenance hotspots, and activity change trends at a glance.

[0039] Furthermore, in order to convert the prediction results into actual management actions, the platform also sets up a maintenance resource scheduling module. This module allocates resources in advance based on the prediction results of maintenance activity, combined with the geographical location, personnel reserves and spare parts inventory of each maintenance outlet. For example, before a certain area is about to enter the high-incidence maintenance period, the platform can notify the corresponding maintenance point in advance to arrange more technicians to station, allocate commonly used spare parts, or inspect and maintain the diagnostic equipment of key models, thereby effectively improving the maintenance response capability, reducing customer waiting time, and reducing service congestion.

[0040] The specific algorithms of graph neural networks and convolutional neural networks are introduced as follows: In the system of the present invention, the graph neural network adopts the form of graph convolutional network (GCN mentioned later) to process the relationship between spatial grids and capture the propagation law of maintenance events in space. The system first divides the urban area into multiple spatial grids, which are regarded as nodes in the graph. If two grids are geographically adjacent, an undirected edge will be established between them, and the adjacency relationship of the entire graph will be constructed in this way. Each node has its own attributes, which are derived from the frequency of various maintenance events in the grid within a specific time window. In other words, the frequency of maintenance events is organized into feature vectors and passed to the GCN model as input data of the node.

[0041] The model structure design of GCN includes two layers of graph convolution. The function of each layer is to update the feature representation of the current node by aggregating the information of neighboring nodes. In detail, the first layer of graph convolution receives the initial features of the node, that is, the feature vector generated based on the frequency of maintenance events, and then generates a set of intermediate feature representations through convolution operations. The second layer of graph convolution then processes these intermediate features, further refines the information, and generates the final node representation for subsequent prediction tasks. The output dimension of each layer of graph convolution is set to 64, which means that the features of each node will be converted into a 64-dimensional vector. In order to enable the model to capture nonlinear relationships, a ReLU activation function is connected after each layer of convolution to enhance the expressive power of the model through this nonlinear transformation. The design of multi-layer graph convolution enables the model to not only focus on the maintenance status of a single grid, but also understand the propagation law of maintenance behavior between adjacent grids, thereby revealing the dynamic trend of maintenance events in space.

[0042] At the same time, the system also uses a one-dimensional convolutional neural network (1D CNN mentioned later) to analyze time series data, with the goal of capturing the changing patterns of maintenance events in the time dimension. The model structure of 1D CNN includes two convolutional layers and one fully connected layer. The first convolutional layer uses a convolution kernel of size 3, a stride of 1, and 32 output channels, which means that it extracts 32 different feature patterns from the input time series. The second convolutional layer also uses a convolution kernel of size 3 and a stride of 1, but the number of output channels is increased to 64, further enriching the expressive power of the features. Each convolutional layer is followed by a ReLU activation function to introduce nonlinear characteristics, and a maximum pooling layer with a pooling window size of 2. This pooling operation can extract local key features in the time series while reducing the dimension of the features and reducing the computational burden. After two layers of convolution and pooling, the extracted features are sent to a fully connected layer and finally mapped to the predicted maintenance activity value, which is a quantitative estimate of future maintenance needs.

[0043] To train these two models, historical maintenance event data from the past year can be used. After these data are sorted, they are divided into training set, validation set, and test set in a ratio of 8:1:1. The training set is used to learn model parameters, the validation set is used to adjust hyperparameters and prevent overfitting, and the test set is used to evaluate the final performance of the model. During training, the batch size is set to 32, which means that 32 samples will be processed simultaneously in each iteration. The entire training process can last for 100 rounds to ensure that the model has enough opportunities to converge to a better solution.

[0044] The loss function of the model can be selected as mean square error to measure the gap between the predicted value and the actual value. The optimization process uses the Adam algorithm, which can dynamically adjust the step size according to the change of the gradient to accelerate convergence. The learning rate is specifically set to 0.001. The performance of the model is evaluated by the root mean square error, and the RMSE on the test set reaches 0.15. This result shows that the average error between the predicted value and the actual value of the model is small, with high prediction accuracy, which can provide a reliable basis for subsequent resource allocation.

[0045] The operation process of the entire algorithm is to combine the analysis of the two dimensions of space and time to form a complete prediction system. First, the system maps the original maintenance event data to the spatial grid and time window to generate a spatiotemporal event matrix. This matrix contains both the frequency of maintenance events for each grid and the sequence of these events over time. Then, based on the geographic adjacency of the spatial grid, the system constructs a graph structure and uses GCN to extract spatial features from it. Through the multi-layer convolution operation of GCN, the model can identify how maintenance events propagate from one grid to adjacent grids, forming dynamic patterns in space.

[0046] At the same time, 1D CNN processes the time series data of each grid. The time series reflects the fluctuation of maintenance events in different time windows. CNN extracts key trends in these fluctuations, such as peaks or periodic patterns, through convolution and pooling operations. Afterwards, the outputs of GCN and CNN are fused together to generate a prediction result for maintenance activity in future time windows. This fusion makes full use of information from both spatial propagation laws and temporal change trends, making the prediction more comprehensive and accurate.

[0047] Finally, the system will optimize the deployment of personnel, equipment, and spare parts based on the predicted maintenance activity and the actual resource situation of the maintenance outlets. For example, if a grid is predicted to be a high-activity area, the system will prioritize allocating more maintenance personnel and spare parts to that area. This prediction-based resource allocation method can significantly improve the response efficiency of maintenance services, reduce waiting time, and improve customer satisfaction.

[0048] Example 1 is as follows: Take City A as an example. Assuming that the city has a permanent population of about 5 million, the number of cars is 2 million, distributed in the main urban area, surrounding urban areas and suburbs. There are about 3,000 auto repair outlets of various types in the city, including fast repair chains, 4S stores and individual repair shops.

[0049] After adopting the solution of the present invention, the platform first deploys a data collection module, accesses the system interface of maintenance outlets throughout the city, and shares the historical traffic event database with the Traffic Management Bureau. The maintenance event data reported by each maintenance point is collected in real time every day, including the time of the event, the latitude and longitude of the vehicle's location, the maintenance type label (such as engine maintenance, tire replacement, brake failure, electrical system, etc.), and some basic vehicle condition information. The collection frequency is in minutes, and tens of thousands of data are collected every day.

[0050] For the convenience of analysis, the platform of the present invention divides the urban area of ​​the entire city A into square grids of 500 meters × 500 meters, forming a total of about 1,800 spatial grids. During the division process, the system refers to the city boundaries, road density and maintenance event heat distribution, while maintaining a unified grid shape and minimizing invalid areas. In terms of time dimension, the platform uses 1 hour as a unit for windowing, so that there are 24 time windows per day.

[0051] When the system receives a new maintenance event data, the data preprocessing module will automatically identify the spatial grid number to which it belongs based on the longitude and latitude of the event, and determine the time window it is in based on the timestamp. Subsequently, the system records the frequency of this type of maintenance event at the corresponding position in the three-dimensional matrix. With the continuous accumulation of data, a stable and dynamically updated spatiotemporal event matrix is ​​gradually formed, covering the distribution statistics of multiple types of maintenance types in the spatial grid and time window of the entire city.

[0052] Next, the platform constructs these spatial grids into a graph structure, where each grid is a node and edges are established between geographically adjacent grids. The system further uses the mature spatiotemporal graph convolutional network model developed in graph neural networks for modeling, which consists of a spatial graph convolution layer and a temporal convolution layer.

[0053] During the training process, the spatial graph convolution layer identifies the transmission trend of maintenance events between different regions through multi-hop information fusion, while the temporal convolution layer identifies the temporal patterns such as growth, decay or periodic fluctuation of maintenance events on nodes.

[0054] After training and verification with a large amount of historical data, the model has strong generalization ability and can make more accurate predictions about future maintenance behaviors.

[0055] For example, if the platform finds through model operation that the frequency of "tire replacement" and "chassis damage" events in several spatial grids in the southwest area of ​​the main urban area has increased significantly in the past three hours, the system predicts that the area may have a peak in maintenance demand in the next two hours, especially concentrated in several areas near highway entrances and exits and intersections with old roads. At the same time, the platform generates a heat map with predicted maintenance activity data, and presents red, orange, yellow and other color levels to distinguish different degrees of activity through a visual interface. Managers can intuitively see the specific scope and intensity of hotspot concentration areas.

[0056] Based on the heat map and model prediction results, the platform automatically triggers the maintenance resource scheduling module and recommends that four maintenance outlets located around the hotspot area prepare in advance: two of the outlets are recommended to increase tire inventory, and the other two are recommended to deploy chassis inspection tools and maintenance technicians. If the user authorizes, the platform can also prompt the driver to avoid high maintenance probability areas or plan conservative routes through the in-vehicle navigation application.

[0057] Throughout the entire process, from the occurrence of maintenance events to data collection, preprocessing, propagation modeling, activity prediction, heat map display and resource scheduling, the platform is based on data-driven and has established a closed-loop intelligent management mechanism to help managers achieve active perception, early intervention and precise allocation, empowering the urban maintenance system and laying the foundation for building a multi-city collaborative maintenance network in the future.

[0058] In summary, the present invention realizes the full-process closed-loop management of automobile aftermarket maintenance behaviors from data perception, trend analysis to resource optimization through the integration of the above-mentioned modeling, prediction, visualization and scheduling mechanisms, which greatly improves the intelligence level and response efficiency of the city-level maintenance service system.

[0059] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent management platform for the automotive aftermarket based on big data, characterized in that: include: The data collection module is used to collect vehicle maintenance event data in multiple areas in real time. Each maintenance event data includes the occurrence time, geographic location information and maintenance type; The space-time grid division module is used to divide the area to be analyzed into multiple space grids according to the preset granularity, and to divide the time continuously into multiple time windows of fixed length; A data preprocessing module, used for mapping the vehicle maintenance event data into corresponding spatial grids and time windows to generate a spatiotemporal event matrix; A maintenance type propagation modeling module is used to analyze the spatiotemporal event matrix, build and train a spatiotemporal graph convolutional network model to simulate the propagation trend of different maintenance types between spatial grids over time; The maintenance activity prediction module is used to predict the maintenance activity of the corresponding maintenance type of each spatial grid at future moments based on the trained spatiotemporal graph convolutional network model.

2. The big data-based automotive aftermarket intelligent management platform according to claim 1, characterized in that: The data collected by the data collection module specifically includes at least the latitude and longitude coordinates corresponding to the maintenance event, a timestamp, and a classification label of the maintenance type.

3. The big data-based automotive aftermarket intelligent management platform according to claim 1 is characterized in that: The space-time grid division module divides the city area into regular square space grids according to the geographical area boundaries of the city map and the preset grid division rules, and each space grid corresponds to a unique space identifier.

4. The big data-based automotive aftermarket intelligent management platform according to claim 3 is characterized in that: The area of ​​a single grid of the spatial grid division rule is between 250 meters × 250 meters and 1 kilometer × 1 kilometer.

5. The big data-based automotive aftermarket intelligent management platform according to claim 1 is characterized in that: The data preprocessing module maps the maintenance event to the corresponding spatial grid and time window according to the latitude and longitude coordinates and timestamp of each maintenance event data, and accumulates and counts the event frequency of the maintenance type in each grid and each time window to generate a spatiotemporal event matrix with multi-dimensional maintenance type statistical values.

6. The big data-based automotive aftermarket intelligent management platform according to claim 1, characterized in that: The maintenance type propagation modeling module establishes a spatiotemporal graph structure, takes spatial grids as graph nodes, and sets edges between adjacent grids for connection; the maintenance type statistics of each node in different time windows are used as node attributes to construct a spatiotemporal graph network structure.

7. The big data-based automotive aftermarket intelligent management platform according to claim 6, characterized in that: The spatiotemporal graph convolutional network model used by the maintenance type propagation modeling module includes a spatial graph convolutional layer and a temporal convolutional layer; The spatial graph convolution layer is used to learn the spatial correlation between nodes in different spatial grids, and the temporal convolution layer is used to capture the trend of each node changing over time.

8. The big data-based automotive aftermarket intelligent management platform according to claim 1 is characterized in that: The maintenance activity prediction module is based on the trained spatiotemporal graph convolutional network model and predicts the maintenance activity values ​​of various maintenance types in each spatial grid in the next time window or multiple consecutive time windows in the future according to the maintenance activity data at the current time and historical time.

9. The big data-based automotive aftermarket intelligent management platform according to claim 8, characterized in that: The maintenance activity value predicted by the maintenance activity prediction module is used to generate a maintenance type heat map, and the heat map intuitively displays the activity level of maintenance types at different spatial grids at future moments.

10. The big data-based automotive aftermarket intelligent management platform according to claim 1, characterized in that: The platform also includes a maintenance resource scheduling module for pre-allocating maintenance personnel, maintenance equipment and spare parts inventory at automobile maintenance outlets based on the prediction results of the maintenance activity prediction module to cope with upcoming high-incidence maintenance areas and maintenance peak periods.

Images