Method for marketing area selection of new energy vehicle based on big data

Abstract

The application provides a new energy automobile marketing area selection method based on big data, selects factors with market characteristics based on vehicle operation data and charging data obtained based on a vehicle networking platform, establishes a marketing hotspot and marketing area recommendation model and a single vehicle monthly portrait, and through combination with map POI point data, can intuitively evaluate the macro market development level, the marketing heat of the local market POI point and the grid area, is convenient for recommending the marketing hotspots and areas to the automobile enterprises, and based on the city user portrait and the area user portrait, supports the automobile enterprises in formulating marketing strategies, can effectively improve the marketing conversion rate, evaluates the marketing effect through the number of newly added vehicles of the vehicle series, and thus forms a closed loop for new energy automobile marketing.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of new energy automobile marketing, and particularly relates to a method for selecting a new energy automobile marketing area based on big data. BACKGROUND

[0002] In recent years, with the breakthrough of key technologies and the maturation of the industrial chain, the new energy automobile industry has developed rapidly, the penetration rate has been continuously improved, and users have gradually accepted new energy automobiles. Some cities have even achieved the goal of 20% of new energy automobile sales in total automobile sales proposed in the New Energy Automobile Industry Development Plan (2021-2035), and the industry development has changed from policy-driven to market-driven.

[0003] At present, the development of new energy automobiles presents an uneven characteristic. From the national perspective, the development of coastal cities is relatively fast, while the development of western cities and northeast regions is relatively slow. From the city perspective, there is a great difference in the number of new energy automobiles in different districts or local regions, reflecting that the market potential of new energy automobiles in different regions is uneven. This brings difficulties to the marketing work of new energy automobiles, and inappropriate selection of marketing areas may lead to sales not reaching the expected target. SUMMARY

[0004] In view of the deficiencies in the prior art, the present application provides a method for selecting a new energy automobile marketing area based on big data, to solve the technical problem that the uneven development of new energy automobiles brings difficulties to the marketing work in the related art.

[0005] The present application provides a method for selecting a new energy automobile marketing area based on big data, comprising:

[0006] Step one, processing the original data of the offline map to determine the basic information of the required POI points;

[0007] Step two, dividing all cities in China using the H3 geographic spatial index system to determine the regular hexagon grid area, and mapping the POI points to a unique regular hexagon grid area;

[0008] Step three, establishing a single behavior feature table for each new energy vehicle according to the running data and static data of the new energy vehicle obtained from the Internet of Vehicles platform, wherein the running data of the new energy vehicle includes a single driving record table and a single charging record table;

[0009] Step four, determining vehicle resident information and vehicle monthly portrait information table according to the single behavior feature table, wherein the vehicle resident information includes a vehicle arrival shopping mall record table and a vehicle resident work and residence record table;

[0010] Step five, according to the vehicle resident information and the single charging record table, respectively determine the POI point and the number of newly added vehicles in the region, the public charging facility coverage information of the administrative region, and determine the city and county user portrait distribution according to the vehicle monthly portrait information;

[0011] Step six, according to the city and county user portrait distribution and single behavior characteristic table, establish the administrative division macro market new energy automobile development level evaluation, local area marketing hotspot and marketing area heat scoring model;

[0012] Step seven, based on the number of newly added vehicles, according to the heat scoring model to evaluate the marketing effect in the hotspot and the region.

[0013] Optionally, the offline map original data is processed to determine the basic information of the required POI point, including:

[0014] The POI point basic information includes POI name, POI type, latitude and longitude, province, city, city code, district and county code, the POI name is standardized, and the standardized name format required by the system is formed;

[0015] According to the POI point basic information, mapping and merging are carried out to obtain three kinds of POI information required by the POI type of shopping mall, residential community and office building.

[0016] Optionally, the H3 geographic spatial index system is used to divide all cities in China to determine the regular hexagon grid area, and the POI point is mapped to a unique regular hexagon grid area, including:

[0017] According to the area of the regular hexagon grid area, the resolution corresponding to the H3 geographic spatial index system is selected, the latitude and longitude of the POI point and the range covered by the regular hexagon grid area are matched to determine the relationship between the POI point and the regular hexagon grid area.

[0018] Optionally, the single behavior characteristic table of each new energy vehicle is established according to the running data and static data of the new energy vehicle obtained from the Internet of vehicles platform, the running data of the new energy vehicle includes single driving record table and single charging record table, including:

[0019] The single driving record table includes vehicle driving start and end time, vehicle parking latitude and longitude, single driving consumption △SOC, single driving time length, single driving mileage;

[0020] The single charging record table includes vehicle charging start and end time, vehicle charging latitude and longitude, single charging time length, fast charging or slow charging;

[0021] Static data, including vehicle model announcement number, vehicle model price, vehicle model power type, vehicle model nominal capacity, vehicle model belonging to vehicle series, vehicle model belonging to company and vehicle model battery type.

[0022] Optionally, the vehicle resident information and the vehicle monthly portrait information table are determined according to the single behavior characteristic table, the vehicle resident information includes a vehicle arrival market record table and a vehicle resident work and residence record table, and the vehicle resident information includes:

[0023] The vehicle resident information includes resident POI points and resident areas, and the POI points and the resident areas are respectively residential communities and office buildings, and the resident POI point and resident area information includes a province, a city, a county, a name, a longitude and a latitude of the resident POI point and the resident area;

[0024] The vehicle monthly portrait information table includes a driving monthly portrait and a charging monthly portrait, and the vehicle monthly portrait information table includes actual endurance mileage, actual 100-kilometer energy consumption, monthly driving days, monthly driving total mileage, daily driving mileage, monthly charging times, monthly fast charging times, monthly slow charging times, fast charging times ratio, average charging time, fast charging average charging time and slow charging average charging time.

[0025] Optionally, the POI points and the regular hexagon grid area new vehicle number, the administrative region public charging facility coverage information are determined according to the vehicle resident information and the single charging record table, and the administrative division user portrait distribution is determined according to the vehicle monthly portrait information table, and the method comprises the following steps:

[0026] The POI points and the regular hexagon grid area new vehicle number are calculated according to the vehicle resident information;

[0027] The administrative region public charging facility coverage information is calculated according to the single charging record table;

[0028] The administrative division user portrait distribution is determined according to the vehicle monthly portrait information table.

[0029] Optionally, the administrative division macro market new energy automobile development level evaluation, the local area marketing hotspot and the marketing area heat score model are established according to the city and county user portrait distribution and the single behavior characteristic table, and the method comprises the following steps:

[0030] According to the city and county user portrait distribution and the single behavior characteristic table, the factor with the market characteristic is selected and the influence weight is given, and the administrative division macro market new energy automobile development level evaluation, the local area marketing hotspot and the marketing area heat score model are constructed.

[0031] Compared with the prior art, the present application has the following beneficial effects:

[0032] The application provides a new energy automobile marketing area selection method based on big data, selects factors with market characteristics based on vehicle operation data and charging data obtained based on a vehicle networking platform, establishes a marketing hotspot and marketing area recommendation model and a single vehicle monthly portrait, and through the combination of map POI point data, can intuitively evaluate the macro market development level, the marketing heat of local market POI points and grid areas, is convenient for recommending marketing hotspots and areas to vehicle enterprises, and based on city user portraits and area user portraits, supports vehicle enterprises in formulating marketing strategies, can effectively improve the marketing conversion rate, evaluates the marketing effect through the number of new vehicles of a vehicle series, so that the new energy automobile marketing forms a closed loop. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 It is a flowchart of the application;

[0034] Figure 2 It is a schematic diagram of the distance between the segmented graph and the adjacent graph in the application;

[0035] Figure 3 It is a schematic diagram of H3 resolution in the application;

[0036] Figure 4 It is a schematic diagram of the marketing hotspot and the marketing area of the embodiment in the application.

[0037] The implementation, functional characteristics and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical scheme and beneficial effects of the application more clear, the technical scheme in the application will be further described below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

[0039] Referring to Figure 1 The application provides a new energy automobile marketing area selection method based on big data, which comprises the following steps:

[0040] Step 1: Process the original data of the offline map to determine the basic information of the required POI points;

[0041] In this embodiment, the required POI point basic information is extracted from the original offline map data, including POI name, POI type, longitude and latitude, province, city, city code, district, district code;

[0042] The original POI type is mapped and merged to obtain a POI point list of the required POI types of shopping malls, residential communities and office buildings;

[0043] The POI name is formatted, including removing abnormal characters, limiting character length, and merging different areas of the same building, to obtain the standardized name format required by the application.

[0044] Step two, using the H3 geospatial indexing system to divide all cities in China, determine the regular hexagon grid area, and map the POI point to a unique regular hexagon grid area;

[0045] In this embodiment, different geospatial segmentation schemes are compared, as shown in the figure, the schemes of triangle, square, and regular hexagon can all achieve full coverage of geospatial without blank parts, but only the scheme of regular hexagon meets the distance from the center point to the adjacent figure is fixed, and the regular hexagon is closer to a circle; Figure 2

[0046] Different resolutions in the H3 geospatial indexing system correspond to regular hexagons of different areas, considering the characteristics of new energy vehicle regional marketing business, the H3 resolution is selected as 7, corresponding to the regular hexagon length of 1.220629759 kilometers, the area of 5.1612932 square kilometers, and each region contains a unique H3 index. Taking Beijing as an example, the urban built-up area is about 1469 square kilometers, containing about 293 regular hexagon grid areas;

[0047] The latitude and longitude of the POI point and the range covered by the regular hexagon area are matched to determine the relationship between the POI point and the grid area, and the corresponding relationship between the two is n:1.

[0048] Step three, according to the running data and static data of new energy vehicles obtained from the Internet of Vehicles platform, a single behavior feature table of each new energy vehicle is established, and the running data of the new energy vehicle includes a single driving record table and a single charging record table;

[0049] In this embodiment, the real running data and static data of new energy vehicles can be obtained from the Internet of Vehicles platform, and a single behavior feature table of each vehicle (VIN is the unique identification of the vehicle) is established according to the running data and static data, which specifically includes:

[0050] The single driving record table R of the vehicle vin includes driving start time driving end time driving duration single driving consumption single driving mileage

[0051] The single parking record table S of the vehicle vin includes parking start time parking end time parking duration​ Parking longitude Parking latitude

[0052] Single charging record table C of the vehicle vin , including charging start time Charging end time Single charging duration Charging longitude Charging latitude

[0053] Static data of the vehicle, i.e. various basic attribute information of the vehicle, including vehicle model announcement number, vehicle model price, vehicle model power type, vehicle model nominal capacity, vehicle model series, and vehicle model company.

[0054] Step four, determining vehicle resident information and vehicle monthly portrait information table according to the single behavior characteristic table, the vehicle resident information including vehicle arrival mall record table and vehicle resident work and residence record table;

[0055] In this embodiment, according to the foregoing single behavior record table, vehicle resident information, whether a private pile user, and a vehicle monthly portrait information table are determined; the vehicle resident information includes resident POI points and resident areas, and the resident POI points and resident areas include a vehicle arrival mall record table and a vehicle resident work and residence record table, and specifically include:

[0056] The vehicle arrival mall record table is determined according to the vehicle stay time in the mall, including the name, longitude and latitude of the vehicle arrival mall belonging to the province, city, county, and the name, longitude and latitude of the mall; the rule for judging the vehicle stay in the mall includes that the vehicle parking position and the mall distance is less than 200 meters, if there are multiple malls within the distance range, the nearest mall is selected, the parking time period is between 9:00-21:00, and the stay duration is greater than or equal to 30 minutes;

[0057] The vehicle resident information table is determined according to the vehicle stay time in the residence and work place, including two dimensions of resident POI points and resident grid areas; the resident information includes the name, longitude and latitude of the resident point and grid area belonging to the province, city, county, and the name, longitude and latitude of the POI point and grid area; the judgment rule of the vehicle resident work place and residence includes that the vehicle monthly stay times are greater than or equal to 10 times or the stay duration is greater than or equal to 80 hours, and the vehicle parking position and the residence or work place distance is less than 200 meters.

[0058] Monthly portrait of driving, including actual range of the vehicle in the current month Actual energy consumption per 100 kilometers Monthly driving days Monthly driving total mileage Daily driving mileage

[0059] Monthly charging profile, including the number of times a user charges per month. Monthly fast charging times Monthly slow charging times Fast charging frequency Average charging time Average charging time per fast charge Average charging time per slow charge

[0060] The formula for calculating the profile metrics is as follows:

[0061] 1. Actual driving range

[0062]

[0063] 2. Actual energy consumption per 100 kilometers

[0064]

[0065] 3. Average daily mileage

[0066]

[0067] 4. Percentage of fast charging cycles

[0068]

[0069] 5. Average charging time per charge

[0070]

[0071] Based on the aforementioned monthly profile metrics, in order to make the profile metrics more stable and prevent abnormal fluctuations from affecting the judgment, the label of each metric is calculated. i The 6-month average is used to obtain the 6-month average portrait information. Calculation formula:

[0072]

[0073] Step 5: Based on the vehicle's permanent location information and the single charging record table, determine the number of new POIs and new vehicles in the region, the coverage information of public charging facilities in the administrative region, and determine the distribution of user profiles in cities and counties based on the vehicle's monthly profile information.

[0074] In this embodiment, the number of newly added vehicles at POI points and in the grid area is calculated based on the vehicle's permanent location information, and the calculation method is as follows:

[0075] Assuming the current month is T0, if a vehicle's permanent location is POI... i Points of Interest (POIs) are not included in the list of vehicle permanent locations from T0-24 month to T0-1 month.i If the point or vehicle is not in the T0-24 month to T0-1 month vehicle residence point table, the vehicle is recorded as a POI point in the current month i The newly added vehicle of the POI point is aggregated to obtain the monthly newly added vehicle number in the POI point and the grid area.

[0076] The administrative region public charging facility coverage information refers to the ratio of the public charging station coverage area to the administrative region built-up area. The public charging station coverage area is, as shown in Figure 3 The H3 geographic spatial index resolution is 8, the corresponding hexagonal grid area is 0.461354684 kilometers, and the area is 0.7373276 square kilometers. If the number of parking times in the current month is greater than or equal to 80 times, it is determined that the administrative region built-up area is determined to be covered by the grid area with the public charging station coverage.

[0077] The city and county user portrait includes user car purchase preferences: vehicle price distribution, power type distribution, positive material distribution, actual range distribution, and actual energy consumption per 100 kilometers distribution; user driving preferences: monthly average driving days distribution, monthly average driving distance distribution, and daily average driving distance distribution; and user charging preferences: monthly average charging times distribution, fast charging times proportion, charging times proportion, average charging time distribution, average charging time distribution (fast charging), and average charging time distribution (slow charging).

[0078] Step six, according to the city and county user portrait distribution and single behavior characteristic table, an administrative division macro market new energy vehicle development level evaluation, local area marketing hotspot and marketing area heat score model are established;

[0079] In this embodiment, according to the vehicle monthly portrait table, an administrative division macro market new energy vehicle development level evaluation and a local area marketing hotspot and marketing area recommendation model are established, which specifically includes:

[0080] The new energy vehicle penetration rate reflects the macro market new energy vehicle development level, which can be measured by the monthly new energy vehicle compulsory insurance premium. The macro market development level is divided into five stages, including the incubation period, the penetration rate is between [0, 10%]; the growth period, the penetration rate is between (10%, 20%]; the development period, the penetration rate is between (20%, 30%]; the mature period, the penetration rate is between (30%, 50%]; and the platform period, the penetration rate is greater than 50%;

[0081] The marketing hotspot and marketing area recommendation model selects the market characteristic factors of private electric passenger car increment proportion, growth level, private charging pile ratio, activity rate, and whether there is public facility coverage to construct a marketing heat evaluation model, including:

[0082] 1. Increment proportion

[0083] Incremental proportion represents the proportion of the number of new energy vehicles added in the POI point or grid area to the number of new vehicles added in the county,

[0084] reflects the proportion of the increment;

[0085] Calculation formula: Incremental proportion = POI point new vehicle number / county new vehicle number

[0086] 2, growth rate

[0087] Growth rate represents the growth rate of new energy vehicles in the POI point or marketing area;

[0088] Calculation formula: Growth rate = (this month's new vehicle number - last month's new vehicle number) / last month's new vehicle number

[0089] 3, private pile ratio

[0090] Private pile ratio represents the proportion of private charging piles owned by vehicle owners in the POI point or grid area. Using private charging piles is a trend in the development of new energy vehicles, and whether the user's area can install private charging piles affects the user's car selection;

[0091] Calculation formula: Private pile ratio = number of private pile running vehicles / total number of running vehicles

[0092] 4, activity rate

[0093] Activity rate represents the frequency of vehicle use. The higher the frequency of vehicle use, the more it can reflect the user's stickiness to the vehicle. At the same time, high active users can increase vehicle exposure and affect the car selection of surrounding users;

[0094] Calculation formula: Activity rate = active running vehicle number / total running vehicle number

[0095] 5, whether there is public facility coverage

[0096] Whether there is public charging facility coverage in the surrounding area will have an important impact on the user's car selection behavior, reflecting the convenience of user charging;

[0097] Calculation formula: if (there is a public charging station within 1000 meters), then whether there is public facility coverage = 1

[0098] Sort and score each marketing factor, and assign an influence weight to establish a comprehensive recommendation heat score model, the calculation formula is as follows:

[0099] Recommendation heat = incremental proportion score*s1+growth rate score*s2+(private pile ratio score+public pile coverage score) / 2*s3+activity rate score*s4

[0100] s iWeight of factor i

[0101] As shown in Figure 4 According to the comprehensive recommendation heat, ranking is performed in the city or district, different recommendation star levels are set, and marketing hotspots and marketing areas with high star levels are recommended to customers.

[0102] Step seven, based on the number of newly added vehicles, the marketing effect in hotspots and areas is evaluated according to the heat scoring model.

[0103] In this embodiment, the marketing effect of the vehicle series is evaluated based on the number of newly added vehicles of the vehicle series, and specifically includes:

[0104] According to the marketing recommendation heat, the vehicle enterprise selects certain marketing hotspots and marketing areas, and according to the user portrait of the city and district, formulates corresponding marketing strategies for specific vehicle series, and carries out marketing work; in order to test the marketing strategy, the number of newly added vehicles of the vehicle series in the corresponding marketing hotspots and marketing areas is analyzed in a monthly and quarterly time cycle, and is compared and ranked with the vehicle series in the same price interval and the same power type in the segmented market, so as to evaluate the marketing effect of the vehicle series, and adjust the marketing strategy as needed.

[0105] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the purpose and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for selecting marketing regions for new energy vehicles based on big data, characterized in that, include: Step 1: Process the raw offline map data to determine the basic information of the required POI points; Step 2: Use the H3 geospatial indexing system to divide all cities in the country, determine the regular hexagonal grid area, and map the POI point to a unique regular hexagonal grid area; Step 3: Based on the operation data and static data of new energy vehicles obtained from the vehicle network platform, establish a single behavior characteristic table for each new energy vehicle. The operation data of the new energy vehicle includes a single driving record table and a single charging record table. Step 4: Based on the single behavior characteristic table, determine the vehicle's permanent location information and the vehicle's monthly profile information table. The vehicle's permanent location information includes the vehicle's arrival at shopping mall record table and the vehicle's permanent work location and residence record table. Step 5: Based on the vehicle's permanent location information and the single charging record table, determine the number of new POIs and new vehicles in the region, the coverage information of public charging facilities in the administrative region, and determine the distribution of user profiles in cities and counties based on the vehicle's monthly profile information. Step 6: Based on the user profile distribution and single behavior characteristic table of the cities and districts, establish an assessment model for the development level of new energy vehicles in the macro market of administrative divisions, a marketing hotspot model for local areas, and a popularity scoring model for marketing regions. Step 7: Based on the number of newly added vehicles, evaluate the marketing effectiveness in hot spots and regions according to the popularity scoring model.

2. The method for selecting marketing regions for new energy vehicles based on big data as described in claim 1, characterized in that, The process of processing the raw offline map data to determine the basic information of the required Points of Interest (POIs) includes: The basic information of the POI includes POI name, POI type, latitude and longitude, province, city, city code, district / county and district / county code. The POI name is standardized to form the standardized name format required by the H3 geospatial index system. Based on the basic information of the POIs, the mapping and merging are performed to obtain the required POI information for three types: shopping malls, residential communities, and office buildings.

3. The method for selecting marketing regions for new energy vehicles based on big data as described in claim 2, characterized in that, The process of using the H3 geospatial indexing system to divide all cities in the country into hexagonal grid regions and mapping each POI point to a unique hexagonal grid region includes: The resolution corresponding to the H3 geospatial indexing system is selected based on the area of ​​the regular hexagonal grid region. The latitude and longitude of the POI point are matched with the range covered by the regular hexagonal grid region to determine the relationship between the POI point and the regular hexagonal grid region.

4. The method for selecting marketing regions for new energy vehicles based on big data as described in claim 1, characterized in that, The process involves establishing a single-transaction behavior characteristic table for each new energy vehicle based on the operational and static data obtained from the vehicle network platform. The operational data for each new energy vehicle includes a single-trip driving record table and a single-charge record table, comprising: The single trip record sheet includes the start and end times of the vehicle trip, the latitude and longitude of the vehicle parking, the △SOC consumed in a single trip, the duration of a single trip, and the mileage of a single trip; The single charging record includes the vehicle charging start and end times, vehicle charging latitude and longitude, single charging duration, and whether it is fast charging or slow charging; Static data includes vehicle announcement number, vehicle price, vehicle power type, vehicle nominal capacity, vehicle series, vehicle company, and vehicle battery type.

5. The method for selecting marketing regions for new energy vehicles based on big data as described in claim 1, characterized in that, The method involves determining vehicle permanent location information and vehicle monthly profile information based on the single behavior feature table. The vehicle permanent location information includes a vehicle arrival record table, a vehicle permanent work location and residence record table, and includes: The vehicle's permanent location information includes the permanent location POI and the permanent location area. The types of POI and permanent location area are residential communities and office buildings, respectively. The permanent location POI and permanent location information include the province, city, district / county, name, longitude, and latitude of the permanent location and location. The vehicle monthly profile information table includes a driving monthly profile and a charging monthly profile, including the vehicle's actual driving range, actual energy consumption per 100 kilometers, number of driving days in the month, total driving mileage in the month, average daily driving mileage, number of charging times in the month, number of fast charging times in the month, number of slow charging times in the month, percentage of fast charging times, average charging time per charging time, average charging time per fast charging time, and average charging time per slow charging time.

6. The method for selecting marketing regions for new energy vehicles based on big data as described in claim 1, characterized in that, The method involves determining the number of newly added vehicles at POI points and in the hexagonal grid area, as well as the coverage information of public charging facilities in administrative regions, based on the vehicle's permanent location information and the single charging record table. It also involves determining the distribution of user profiles by administrative division based on the monthly vehicle profile information table, including: Based on the vehicle resident information, calculate the number of newly added vehicles at POI points and in the regular hexagonal grid area; Based on the single charging record table, calculate the coverage information of public charging facilities in the administrative region; Based on the monthly vehicle profile information table, the distribution of user profiles by administrative region is determined.

7. The method for selecting marketing regions for new energy vehicles based on big data as described in claim 1, characterized in that, The process involves establishing an assessment model for the development level of new energy vehicles in the macro-market of administrative divisions, a model for marketing hotspots in local areas, and a model for scoring the popularity of marketing regions, based on the distribution of user profiles and single-transaction behavior characteristics of cities and counties. Based on the distribution of user profiles and single-time behavior characteristics in cities and counties, factors with market characteristics are selected and assigned influence weights to construct an assessment model of the development level of new energy vehicles in the macro market of administrative divisions, a marketing hotspot model in local areas, and a popularity scoring model for marketing regions.

Images