Data analysis method and system based on vehicle-mounted terminal

Abstract

The invention relates to the technical field of intelligent site selection, and particularly discloses a data analysis method and system based on a vehicle-mounted terminal, and the method comprises the following steps: obtaining the coordinates of a vehicle during parking based on the vehicle-mounted terminal in a preset after-sales area, and carrying out the clustering of the coordinates, and obtaining a cluster; obtaining a minimum circumcircle of the clustering cluster, marking the minimum circumcircle as a screening circle, obtaining the position of the after-sales service station in the screening circle, and correcting the clustering cluster according to the position of the after-sales service station to obtain a target cluster; and obtaining a to-be-selected position, and selecting a new after-sales service station position in combination with the theoretical center of gravity of the target cluster. According to the invention, the site selection layout of the after-sales service station can be optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent site selection, and in particular to a data analysis method and system based on a vehicle-mounted terminal. Background Art

[0002] The on-board terminal is a type of intelligent device installed inside the vehicle. It is used to collect, process and transmit vehicle and road information in real time, such as location, speed and vehicle status, and provide remote monitoring and other services to ensure driving safety and traffic efficiency.

[0003] When a vehicle is undergoing maintenance or a breakdown occurs, the owner usually needs to go to a fixed after-sales service station for repair or inspection. The geographical location of the after-sales service station is very important to the car owner. Choosing a suitable location to set up an after-sales station can effectively shorten the distance and time for the car owner to go for repair, thereby significantly improving the convenience of the car owner, which can not only improve customer satisfaction with after-sales service, but also help to improve the overall reputation and market competitiveness of the brand.

[0004] In the existing technology, after-sales service stations are usually set up based on manual experience. This method is too subjective and lacks scientific and reasonable evaluation basis. It may lead to unbalanced station layout, resulting in idle after-sales service resources in some areas and insufficient service stations in other areas. This will not only fail to effectively meet the actual needs of car owners, but also reduce customer experience and satisfaction, and thus have an adverse impact on the company's brand image and market competitiveness. Summary of the invention

[0005] The purpose of the present invention is to provide a data analysis method and system based on a vehicle-mounted terminal to solve the following technical problems: In the existing technology, after-sales service stations are usually set up based on manual experience. This method is too subjective and lacks scientific and reasonable evaluation basis. It may lead to unbalanced station layout, resulting in idle after-sales service resources in some areas and insufficient service stations in other areas. This will not only fail to effectively meet the actual needs of car owners, but also reduce customer experience and satisfaction, and thus have an adverse impact on the company's brand image and market competitiveness.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A data analysis method based on a vehicle terminal comprises the following steps: In a preset after-sales area, the coordinates of the vehicle when it is parked are obtained based on the vehicle terminal, and the coordinates are clustered to obtain clusters; Obtaining the minimum circumscribed circle of the cluster, recorded as a screening circle, obtaining the location of the after-sales service station within the screening circle, and correcting the cluster according to the location of the after-sales service station to obtain the target cluster; Obtain the candidate positions of after-sales service stations within the target cluster, obtain the theoretical centroid x of the target cluster, obtain the candidate position X closest to the theoretical centroid x, obtain the areas of the triangles formed by any two after-sales service stations and the candidate position X. If the largest triangle area S≥0.8Y, then use the candidate position X as the new after-sales service station position, where Y = min(y1, y2), y1 and y2 respectively represent the area of the smallest circumscribed circle and the area of the smallest circumscribed rectangle of the target cluster, and min(y1, y2) represents the minimum value selected from y1 and y2.

[0007] As a further solution of the present invention: The process of clustering the coordinates specifically includes: Set the first radius d, with the coordinate a as the center, calculate the coordinate density within the first radius d, set the first density threshold fys. If the coordinate density is greater than or equal to the first density threshold, then with the coordinate a as the center, generate an initial cluster A with a clustering radius of d, obtain the theoretical centroid A1 of the initial cluster A, and use the coordinate density within the clustering radius d as the weight of the theoretical centroid A1. Set the second radius E, with the theoretical centroid i as the center, calculate the judgment density within the second radius , where J represents the total number of theoretical centroids within the second radius E, and Cj represents the weight of the j-th theoretical centroid within the second radius E. If the judgment density H≥Fys, with the theoretical centroid i as the center, generate a clustering cluster with a clustering radius of E, where Fys represents the preset second density threshold, and Fys>fys.

[0008] As a further solution of the present invention: During the process of generating the clustering cluster, the following steps are further included: If the judgment density H<Fys, then mark the theoretical centroids within the second radius E as undetermined centroids; With the undetermined centroid as the center, calculate the coordinate density within the first radius d, denoted as the undetermined density. Within the preset first radius range , increase the first radius d by the preset first radius correction value d1, where η is a preset coefficient and η>1, and obtain the new undetermined density after each increase in the first radius; Obtain the maximum undetermined density of a single undetermined centroid. If the maximum undetermined density is greater than or equal to 3fys, obtain the first radius D corresponding to the maximum undetermined density, and with the undetermined centroid as the center, generate a clustering cluster with a clustering radius of D.

[0009] As a further solution of the present invention: The process of correcting the clustering cluster according to the after-sales service station position to obtain the target cluster specifically includes: Set the screening radius R, with the after-sales service station position as the center, calculate the coordinate density P within the screening radius R; A coordinate density threshold value Pys is set. If the coordinate density P>Pys, the screening radius R is slowly increased at a preset speed until the coordinate density P1 in the corresponding screening circle is equal to Pys after a certain adjustment of the screening radius; if the coordinate density P<Pys, the screening radius R is slowly reduced at the speed until the coordinate density P1 in the corresponding screening circle is equal to Pys after a certain adjustment of the screening radius; When the coordinate density within the screening circle is equal to Pys, the coordinates within the screening circle are removed to obtain a new cluster, which is recorded as the target cluster.

[0010] As a further solution of the present invention: after the candidate location X is used as the new after-sales service station location, the following steps are also included: Set the maximum triangle area threshold Sys. If the maximum triangle area S ≥ Sys, perform the following steps: Sorting the triangle areas in descending order to obtain a first sorting, starting from the area of ​​the first triangle in the first sorting, determining the first triangle area smaller than Sys, recording it as the target area, taking the triangle corresponding to the target area as the target triangle, removing the coordinate points in the target triangle, and obtaining a new target cluster Z; The theoretical center of gravity z of the target cluster Z is obtained, and the new after-sales service station position is determined again based on the theoretical center of gravity z.

[0011] As a further solution of the present invention: if all the candidate locations are selected as new after-sales service station locations, information is sent for reporting and selection is performed manually.

[0012] As a further solution of the present invention: if there is no after-sales service station in a certain target cluster, the candidate position that is closest to the theoretical center of gravity of the target cluster is used as the new after-sales service station position.

[0013] A data analysis system based on a vehicle terminal, comprising: Clustering module: in a preset after-sales area, based on the vehicle terminal, the coordinates of the vehicle when it is parked are obtained, and the coordinates are clustered to obtain clusters; Cluster optimization module: obtaining the minimum circumscribed circle of the cluster cluster, recorded as the screening circle, obtaining the location of the after-sales service station within the screening circle, and correcting the cluster cluster according to the location of the after-sales service station to obtain the target cluster; Selection module: Obtain the candidate locations of after-sales service stations within the target cluster, obtain the theoretical centroid x of the target cluster, obtain the candidate location X closest to the theoretical centroid x, obtain the areas of the triangles formed by any two after-sales service stations and the candidate location X. If the largest triangle area S ≥ 0.8Y, then use the candidate location X as the new location of the after-sales service station, where Y = min(y1, y2), y1 and y2 respectively represent the minimum circumscribed circle area and the minimum circumscribed rectangle area of the target cluster, and min(y1, y2) represents the minimum value selected from y1 and y2.

[0014] Advantages of the present invention: Compared with the prior art, in the present invention, firstly, the vehicle parking coordinates are preliminarily clustered by setting the first radius d and the first density threshold fys to form an initial cluster and calculate the theoretical centroid A1. The advantage of using the coordinate density as the centroid weight is that it can identify and lock the hot spots in advance in the case of a large amount of data and a relatively dispersed distribution, thus greatly improving the efficiency and accuracy of subsequent clustering operations. Then, based on the second radius E and the second density threshold Fys, by calculating and judging the density H and comparing it with Fys, if H ≥ Fys, a new clustering cluster is generated, which can effectively eliminate noise points and perform more refined scale control on the clustering radius. If H < Fys, these centroids are marked as undetermined centroids and in the subsequent radius expansion range The first radius d is gradually increased, and it is repeatedly adjusted by the preset first radius correction value d1 to find the maximum pending density threshold that can make the pending center of gravity meet 3fys. When the threshold is found, the gradient change of the actual data distribution can be used to divide the effective clustering clusters that best represent the local hotspots. This step-by-step correction method reduces the risk of rough or excessive clustering divisions caused by a one-time fixed radius, thereby making the target cluster closer to the actual distribution of vehicles; then the minimum circumscribed circle of the cluster is obtained and called the screening circle, and the cluster is corrected in combination with the existing after-sales service station location. By setting the screening radius R and comparing it with the coordinate density threshold Pys, the strategy of slowly increasing or decreasing the screening radius R is determined. This on-demand adjustment mechanism can ensure that the extracted target cluster is kept at a balanced density level while avoiding the deviation caused by the subjective radius specified by humans, thereby ensuring that the eliminated coordinate points accurately point to those areas that are not very meaningful in the after-sales service layout; when the target cluster is In further analysis, based on the candidate locations of the after-sales service stations in the target cluster and the candidate location X that is closest to the theoretical center of gravity x, by calculating the triangular area formed by any two after-sales service stations related to the location, if the maximum triangular area S satisfies S≥0.8Y, X is directly used as the new after-sales service station location, ensuring that the site selection of the new station has higher stability and discrimination while meeting the coverage requirements; next, the maximum triangular area threshold Sys is introduced, and all triangular areas are arranged in descending order and compared one by one. When the first area smaller than Sys is found, the coordinate points in the corresponding triangle are eliminated to obtain a new target cluster Z, and the theoretical center of gravity z of the target cluster Z is obtained again, and the new after-sales service station location is corrected or determined accordingly, which helps to achieve a better network layout arrangement for the entire after-sales area after excluding coordinate points that have insufficient contribution to the service layout or are distributed on the edge, providing accurate data support and algorithm support for ultimately reducing after-sales costs, improving service accessibility and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below in conjunction with the accompanying drawings.

[0016] Figure 1 It is a flow chart of a data analysis method based on a vehicle-mounted terminal of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Please refer to Figure 1 As shown, the present invention is a data analysis method based on an in-vehicle terminal, including the following steps: Within a preset after-sales area, based on the in-vehicle terminal, obtain the coordinates of the vehicle when it is parked, and cluster the coordinates to obtain a clustering cluster; In a preferred embodiment of the present invention, the process of clustering the coordinates specifically includes: Set the first radius d, with the coordinate a as the center, calculate the coordinate density within the first radius d, set the first density threshold fys, if the coordinate density is greater than or equal to the first density threshold, then with the coordinate a as the center, generate an initial cluster A with a clustering radius of d, obtain the theoretical center of gravity A1 of the initial cluster A, and use the coordinate density within the clustering radius d as the weight of the theoretical center of gravity A1; Set the second radius E, with the theoretical center of gravity i as the center, calculate the judgment density within the second radius , J represents the total number of theoretical centers of gravity within the second radius E, Cj represents the weight of the jth theoretical center of gravity within the second radius E, if the judgment density H≥Fys, with the theoretical center of gravity i as the center, generate a clustering cluster with a clustering radius of E, Fys represents a preset second density threshold, and Fys>fys.

[0019] It can be understood that, with the user's authorization, based on the in-vehicle terminal, obtain the coordinates of the user when parking. The coordinates when parking reflect relatively common locations such as the user's living address, and use this as the data basis for subsequent layout of service point locations; set the first radius d based on requirements or empirical data, which is used to measure the distribution of a certain coordinate within a certain range around it. Then, with the coordinate a as the center, calculate the coordinate density within the first radius d, which can be characterized by a simple counting method. For example, count how many vehicles (or how many parking coordinates) fall within this circular area within a distance not exceeding d from the coordinate a, and this value can be called the coordinate density; set the first density threshold fys: it is used to measure whether there are enough coordinate points within a small range. If it exceeds this threshold, it means that the area around the coordinate a is relatively crowded, and an initial clustering cluster can be generated here. For example, if fys = 30, it means that if the number of vehicles within the radius d≥30, the condition for establishing an initial cluster is met; Similar to the first radius d, the value of E is usually set according to project requirements, experience, or experimental data. It is used to measure the degree of aggregation of a certain theoretical center of gravity within a larger range. Then, with the theoretical center of gravity i as the center, calculate the judgment density within the second radius E: Assume that currently, we regard the theoretical centers of gravity A1, A2, A3... of each initial cluster as a "set of center of gravity points". Then, with any one of the theoretical centers of gravity i as the center, count how many other theoretical centers of gravity fall within the range of radius E; at the same time, each theoretical center of gravity corresponds to its own weight (for example, the total coordinate density of the initial cluster corresponding to the center of gravity A1). Sum up the weights of all the centers of gravity that fall within E, and the resulting value is the judgment density H. Since the judgment object here has become the theoretical center of gravity itself, the threshold Fys is generally higher than fys to ensure that only when the density of the centers of gravity is high enough, it is considered that this area has the basis for merging to form a larger cluster; if the judgment density H ≥ Fys, then generate a cluster with a cluster radius of E centered on the theoretical center of gravity i: This means that if there have been accumulated enough weights of other centers of gravity in the area with a radius of E near the center of gravity i, then the coordinates (not just the centers of gravity) within this range can be merged into a larger cluster and regarded as a candidate area or the final effective cluster; In a preferred case of this embodiment, if the judgment density H < Fys, then mark the theoretical centers of gravity within the second radius E as undetermined centers of gravity; With the undetermined center of gravity as the center, calculate the coordinate density within the first radius d, denoted as the undetermined density, within the preset first radius range Increase the first radius d by the preset first radius correction value d1 within the range, where η is a preset coefficient and η > 1, and obtain a new undetermined density each time the first radius is increased; Obtain the maximum undetermined density of a single undetermined center of gravity. If the maximum undetermined density is greater than or equal to 3fys, obtain the first radius D corresponding to the maximum undetermined density, and generate a cluster with a cluster radius of D centered on the undetermined center of gravity; It should be noted that if the calculated judgment density H is less than Fys, a cluster will not be generated immediately. Instead, all the theoretical centers of gravity within the second radius E will be marked as undetermined centers of gravity; in this case, with any undetermined center of gravity as the center, count the number of coordinates within its first radius d again to calculate the undetermined density, and within the preset first radius range Increase d multiple times within, with each increase amplitude being the first radius correction value d1. For example, when the initial d = 1 km, η = 2, and d1 = 0.2 km, the coordinate density statistics of the same undetermined centroid will be carried out at radii such as 1 km, 1.2 km, 1.4 km, 1.6 km, 1.8 km, 2 km, etc., so as to obtain a set of undetermined densities that continuously expand with the radius. Finally, select the maximum undetermined density value that a single undetermined centroid can reach. If this maximum undetermined density exceeds 3 fys, it indicates that at an appropriately enlarged radius, the coordinate distribution around this undetermined centroid has a sufficiently high density. At this time, the first radius D corresponding to the maximum undetermined density can be obtained, and a new clustering cluster is generated with this undetermined centroid as the center and D as the clustering radius, which can further explore potential high-density regions in the case of initial H < Fys and avoid ignoring the deep aggregation phenomenon existing locally due to single determination. Obtain the minimum circumscribed circle of the clustering cluster obtained above, denoted as the screening circle, and obtain the positions of after-sales service stations within the screening circle, and correct the clustering cluster according to the positions of after-sales service stations to obtain the target cluster. A preferred case in this embodiment: The process of correcting the clustering cluster according to the positions of after-sales service stations to obtain the target cluster specifically includes: Set a screening radius R, and calculate the coordinate density P within the screening radius R with the position of the after-sales service station as the center. Set a coordinate density threshold Pys. If the coordinate density P > Pys, then slowly increase the screening radius R at a preset speed until after a certain adjustment of the screening radius, the coordinate density P1 within the corresponding screening circle = Pys. If the coordinate density P < Pys, then slowly decrease the screening radius R at the above speed until after a certain adjustment of the screening radius, the coordinate density P1 within the corresponding screening circle = Pys. When the coordinate density within the screening circle is equal to Pys, remove the coordinates within the screening circle to obtain a new clustering cluster, denoted as the target cluster. It should be noted that an initial screening radius R is set and the coordinate density P within the circumference is calculated with the location of the after-sales service station as the center. Subsequently, in order to adapt to the balance between coordinate distribution and after-sales service needs, a coordinate density threshold Pys is introduced. If the current coordinate density P is greater than Pys, the boundary is gradually expanded outward by increasing the screening radius R by a small amount and multiple times until the coordinate density P1 obtained after a certain radius adjustment just drops to Pys. On the contrary, if P is less than Pys, the screening radius R is gradually reduced by the same means, so that the coordinate density P1 within the screening circle is gradually increased to Pys. When the coordinate density within the screening circle is equal to Pys, the coordinate points covered by the screening circle are removed to obtain a new, more focused cluster as the target cluster. This dynamic radius adjustment approach can ensure that only coordinate points matching the predetermined density threshold are retained around the service station, thereby better filtering discrete data while providing a more reasonable regional division basis for the layout and site selection of subsequent after-sales service stations. Obtain the candidate position of the after-sales service station in the target cluster, obtain the theoretical center of gravity x of the target cluster, obtain the candidate position X that is closest to the theoretical center of gravity x, obtain the area of ​​the triangle formed by any two after-sales service stations and the candidate position X, if the maximum triangle area S ≥ 0.8Y, then take the candidate position X as the new after-sales service station position, Y = min (y1, y2), y1, y2 represent the minimum circumscribed circle area and the minimum circumscribed rectangle area of ​​the target cluster respectively, min (y1, y2) represents the minimum value of y1 and y2; It is worth noting that in order to further optimize the layout of the service stations, the candidate locations of the existing after-sales service stations will be obtained in the obtained target cluster, and the theoretical center of gravity x of the target cluster will be calculated. Then, a candidate location X that is closest to the theoretical center of gravity x will be selected; then, the candidate location X will be connected with any two existing after-sales service stations in the target cluster to form a triangle, and the areas of all possible triangles will be calculated by alternating combinations, and the largest triangle area S will be found among them; if the largest triangle area S satisfies the condition S≥0.8Y, then the candidate location X will be directly connected to the target cluster. Once the new after-sales service station location is confirmed, the station can take into account the coverage and service efficiency of the surrounding areas; the value of Y is the smaller value between y1 and y2 (Y=min(y1, y2)), y1 represents the minimum circumscribed circle area of ​​the target cluster, and y2 represents the minimum circumscribed rectangle area of ​​the target cluster. The original intention of doing so is to use the most compact outer geometric shape to constrain the feasibility range of the station triangle layout, so as to avoid the waste of resources caused by the selection of new stations too randomly or deviating from the center of the cluster, and finally form an after-sales service station network with good compatibility and reasonable density distribution; In another preferred embodiment of the present invention, after the candidate location X is used as the new after-sales service station location, the following steps are also included: Set the maximum triangle area threshold Sys. If the maximum triangle area S ≥ Sys, perform the following steps: Sorting the triangle areas in descending order to obtain a first sorting, starting from the area of ​​the first triangle in the first sorting, determining the first triangle area smaller than Sys, recording it as the target area, taking the triangle corresponding to the target area as the target triangle, removing the coordinate points in the target triangle, and obtaining a new target cluster Z; The theoretical center of gravity z of the target cluster Z is obtained, and the new after-sales service station position is determined again based on the theoretical center of gravity z.

[0020] It is understandable that if the candidate location X has been confirmed as the new after-sales service station location, but the triangle area is still large, further fine-tuning can be done: first, a maximum triangle area threshold Sys is set. When the calculated maximum triangle area S is greater than or equal to Sys, all triangle areas are sorted in descending order, and starting from the first place in the sorting result, the first triangle area smaller than Sys is found as the target area. The triangle corresponding to the target area is then regarded as the target triangle and the coordinate points within the target triangle range are removed to form a new target cluster Z. Subsequently, the theoretical center of gravity z of this target cluster Z is continued to be calculated, and the appropriate after-sales service station location is determined again with reference to the center of gravity. In this way, the synergy between the existing site and the new site can be fully utilized, and local splitting can be performed in the area with too large an area, so as to ensure that the overall service distribution is more reasonable and more in line with actual needs. In a preferred situation of this embodiment, if all the candidate locations are selected as new after-sales service station locations, information is sent for reporting and selection is performed manually.

[0021] In another preferred embodiment of the present invention, if there is no after-sales service station in a certain target cluster, the candidate position closest to the theoretical center of gravity of the target cluster is used as the new after-sales service station position.

[0022] A data analysis system based on a vehicle terminal, comprising: Clustering module: in a preset after-sales area, based on the vehicle terminal, the coordinates of the vehicle when it is parked are obtained, and the coordinates are clustered to obtain clusters; Cluster optimization module: obtaining the minimum circumscribed circle of the cluster cluster, recorded as the screening circle, obtaining the location of the after-sales service station within the screening circle, and correcting the cluster cluster according to the location of the after-sales service station to obtain the target cluster; Selection module: obtain the candidate position of the after-sales service station in the target cluster, obtain the theoretical center of gravity x of the target cluster, obtain the candidate position X that is closest to the theoretical center of gravity x, obtain the area of ​​the triangle formed by any two after-sales service stations and the candidate position X, if the maximum triangle area S≥0.8Y, then the candidate position X is used as the new after-sales service station position, Y=min(y1, y2), y1 and y2 represent the minimum circumscribed circle area and the minimum circumscribed rectangle area of ​​the target cluster respectively, and min(y1, y2) represents selecting the minimum value of y1 and y2.

[0023] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A data analysis method based on an in-vehicle terminal, characterized in that It includes the following steps: Within a preset after-sales area, based on the in-vehicle terminal, obtain the coordinates of the vehicle when it is parked, and cluster the coordinates to obtain a clustering cluster; Obtain the minimum circumscribed circle of the clustering cluster, denoted as the screening circle, obtain the locations of after-sales service stations within the screening circle, and correct the clustering cluster according to the locations of after-sales service stations to obtain a target cluster; Obtain the candidate locations of after-sales service stations within the target cluster, obtain the theoretical centroid x of the target cluster, obtain the candidate location X closest to the theoretical centroid x, obtain the areas of the triangles formed by any two after-sales service stations and the candidate location X. If the maximum triangle area S≥0.8Y, then use the candidate location X as the new location of the after-sales service station, where Y = min(y1, y2), and y1 and y2 respectively represent the area of the minimum circumscribed circle and the area of the minimum circumscribed rectangle of the target cluster, and min(y1, y2) represents the minimum value selected from y1 and y2.

2. The data analysis method based on a vehicle-mounted terminal according to claim 1, wherein The process of clustering the coordinates specifically includes: Set the first radius d, with the coordinate a as the center, calculate the coordinate density within the first radius d, set the first density threshold fys. If the coordinate density is greater than or equal to the first density threshold, then with the coordinate a as the center, generate an initial cluster A with a clustering radius of d, obtain the theoretical centroid A1 of the initial cluster A, and use the coordinate density within the clustering radius d as the weight of the theoretical centroid A1; Set a second radius E, centered on the theoretical center of gravity i, and calculate the judgment density within the second radius , J represents the total number of theoretical centers of gravity within the second radius E, Cj represents the weight of the j-th theoretical center of gravity within the second radius E. If the judgment density H≥Fys, a clustering cluster with a clustering radius of E is generated centered on the theoretical center of gravity i. Fys represents a preset second density threshold, and Fys>fys.

3. The data analysis method based on a vehicle-mounted terminal according to claim 2, wherein During the process of generating the clustering cluster, it also includes the following steps: If the judgment density H < Fys, then mark the theoretical centroid within the second radius E as the to-be-determined centroid; With the to-be-determined center of gravity as the center, calculate the coordinate density within the first radius d, denoted as the to-be-determined density. Within the preset first radius range increase the first radius d by a preset first radius correction value d1. η is a preset coefficient and η > 1, and obtain a new to-be-determined density after each increase in the first radius; Obtain the maximum to-be-determined density of a single to-be-determined centroid. If the maximum to-be-determined density is greater than or equal to 3fys, obtain the first radius D corresponding to the maximum to-be-determined density, and with the to-be-determined centroid as the center, generate a clustering cluster with a clustering radius of D.

4. A data analysis method based on an in-vehicle terminal according to claim 1, characterized in that, The process of correcting the clustering cluster according to the locations of after-sales service stations to obtain the target cluster specifically includes: Set the screening radius R, with the location of the after-sales service station as the center, calculate the coordinate density P within the screening radius R; Set the coordinate density threshold Pys. If the coordinate density P > Pys, then slowly increase the screening radius R at a preset speed until after a certain adjustment of the screening radius, the coordinate density P1 within the corresponding screening circle = Pys; if the coordinate density P < Pys, then slowly decrease the screening radius R at the speed until after a certain adjustment of the screening radius, the coordinate density P1 within the corresponding screening circle = Pys; When the coordinate density within the screening circle is equal to Pys, remove the coordinates within the screening circle to obtain a new clustering cluster, denoted as the target cluster.

5. The data analysis method based on a vehicle-mounted terminal according to claim 1, wherein After using the candidate location X as the new location of the after-sales service station, it also includes the following steps: Set the maximum triangle area threshold Sys. If the maximum triangle area S≥Sys, execute the following steps: Sort the triangle areas in descending order to obtain the first sorting. Starting from the triangle area at the top of the first sorting, determine the first triangle area less than Sys, denoted as the target area. Take the triangle corresponding to the target area as the target triangle, and remove the coordinate points within the target triangle to obtain a new target cluster Z; Obtain the theoretical centroid z of the target cluster Z, and determine the location of the new after-sales service station again based on the theoretical centroid z.

6. The data analysis method based on a vehicle-mounted terminal according to claim 5, wherein, If all candidate locations are used as the locations of the new after-sales service stations, send information for reporting and make a selection manually.

7. A data analysis method based on an in-vehicle terminal according to claim 1, characterized in that If there is no after-sales service station in a certain target cluster, the candidate location closest to the theoretical centroid of the target cluster is used as the location of the new after-sales service station.

8. A data analysis system based on an in-vehicle terminal, characterized in that, It includes: Clustering module: In the preset after-sales area, based on the in-vehicle terminal, obtain the coordinates when the vehicle parks, and cluster the coordinates to obtain clustering clusters; Clustering optimization module: Obtain the minimum circumscribed circle of the clustering clusters, denoted as the screening circle, obtain the locations of the after-sales service stations within the screening circle, and correct the clustering clusters according to the locations of the after-sales service stations to obtain target clusters; Selection module: Obtain the candidate locations of the after-sales service stations within the target cluster, obtain the theoretical centroid x of the target cluster, obtain the candidate location X closest to the theoretical centroid x, obtain the areas of the triangles formed by any two after-sales service stations and the candidate location X. If the largest triangle area S≥0.8Y, then use the candidate location X as the location of the new after-sales service station, where Y = min(y1, y2), and y1 and y2 respectively represent the area of the minimum circumscribed circle and the area of the minimum circumscribed rectangle of the target cluster, and min(y1, y2) represents the minimum value selected from y1 and y2.

Images