Insurance customer resource optimal configuration method and system based on position calculation

Abstract

The invention provides an insurance customer resource optimal configuration method and system based on position calculation, and the method comprises the steps: selecting a visiting position, the occurrence frequency of which is greater than a preset first frequency threshold value in a preset first period from the last visiting time, and calculating a first candidate calculation position through a coordinate center point weighting calculation method, determining an actual contactable position of the to-be-distributed insurance customer; selecting visit positions of all insurance customers corresponding to the salesman within a preset second period from the last visit time, calculating a second candidate calculation position through a coordinate center point weighted calculation method, and determining a time sequence activity area center of the to-be-distributed salesman; according to the actual contactable position of the to-be-distributed insurance customer and the time sequence activity area center of the to-be-distributed salesman, weighting and calculating a distance parameter; and allocating the to-be-allocated insurance customer to the to-be-allocated salesman according to the distance parameter after the weighted calculation, thereby improving the reliability of the resource optimization configuration of the insurance customer.

Description

Technical Field

[0001] This invention relates to the field of insurance customer resource allocation, and in particular to a method and system for optimizing the allocation of insurance customer resources based on location computing. Background Technology

[0002] In the life insurance industry, agent turnover often results in a large number of customers needing to be reassigned to other agents to maintain service continuity. Traditional customer resource allocation methods include two approaches: one is based on "sales team kinship" to assign customers to the supervisors of departing agents, and the other is based on "customer's registered address" to agents in nearby sales offices.

[0003] However, the allocation method based on the "blood relationship of the sales team" in the relevant technologies only considers the management link and does not consider whether the policyholders of the off-site salespersons are within reach of their supervisors. While the allocation method based on the "customer's registered address" takes into account the location relationship between the customer and the salesperson, it lacks verification of the authenticity of the customer's address. According to statistics, in 2020, 55% of the urban population in China lived separately from their registered address, of which 42% were migrant workers. Therefore, the probability of a customer's policy registration address changing in the future is very high. In fact, some customers' registered addresses are not reachable addresses, such as the customer's workplace not in the local area.

[0004] Therefore, both traditional methods ignore the actual reach of customers and the activity range of salespersons. It is very likely that customer resources will be mismatched because the customer visit distance is too far or the salesperson is not active in the customer's area. This will affect the timely access of customers to services and may even result in the loss of some customers' opportunities for secondary development, which is not conducive to improving the reliability of the optimal allocation of insurance customer resources.

[0005] To address the issue of low reliability in the optimization and allocation of insurance customer resources in existing technologies, this solution is proposed. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention innovatively proposes a location-based computational method for optimizing the allocation of insurance customer resources. This method effectively solves the problem of low reliability in optimizing the allocation of insurance customer resources caused by existing technologies, and effectively improves the reliability of optimizing the allocation of insurance customer resources.

[0007] The first aspect of this invention provides a method for optimizing the allocation of insurance customer resources based on location calculation, comprising: Obtain visit history information of insurance customers to be assigned and agents to be assigned; the historical information includes visit location history information and visit time history information. Insurance customers to be assigned are policy customers whose original agents have left the company, or existing policy customers whose addresses have changed and who have not registered the new addresses with the insurance company. Based on the historical information of the visit location and visit time of the insurance customers to be assigned, the visit location that appears more frequently than a preset first frequency threshold within a preset first cycle since the last visit time is selected as the first candidate calculation location. The first candidate calculation location is calculated using the coordinate center point weighted calculation method to determine the actual accessible location of the insurance customers to be assigned. Based on the historical information of the visit locations and visit times of the salesperson to be assigned, the visit locations of all insurance customers of the salesperson within the second cycle from the last visit time are selected as the second candidate calculation locations. The second candidate calculation locations are calculated using the coordinate center point weighted calculation method to determine the center of the time sequence activity area of ​​the salesperson to be assigned. Based on the actual accessible location of the insurance customers to be assigned and the center of the activity area of ​​the agents to be assigned, a weighted distance parameter is calculated; based on the weighted distance parameter, the insurance customers to be assigned are assigned to the agents to be assigned.

[0008] Optionally, selecting a visit location whose frequency of occurrence within a preset first period after the last visit is greater than a preset first frequency threshold as the first candidate calculation location specifically includes: Based on the visit location information of all agents corresponding to the insurance customers to be assigned, using the application to record customer visits, construct a set of the customer's first coordinate location; In the first set of customer coordinate locations, based on the visit time information of the customer visit, the first preset period is deduced by going back from the visit time of the last insurance customer to be assigned, and all coordinate locations recorded by the application within the first preset period are selected to construct the second set of customer coordinate locations. Select the coordinates in the second set of customer coordinates that appear more frequently than a first preset percentage of the total number of visits to the insurance customers to be assigned, and use them as the first candidate calculation locations; where the total number of visits is the number of visits confirmed by the insurance customers to be assigned on-site by scanning the code or the visit duration information is greater than a preset first duration threshold.

[0009] Furthermore, selecting visit locations whose frequency exceeds a preset first frequency threshold within a preset first period from the last visit time as the first candidate calculation locations specifically includes: If there is no coordinate location in the customer coordinate location set that has a frequency greater than the first preset percentage of the total number of visits to the insurance customers to be assigned within the first preset period, then select the preset first number of coordinate locations in the customer coordinate location set that are closest to the last visit time in chronological order. If there are no visit records for the insurance customer to be assigned, the address registered on the customer's most recent policy will be converted into location coordinates and used as the first candidate location for calculation.

[0010] Optionally, the first candidate calculation location is calculated using a weighted calculation method based on the coordinate center point to determine the actual accessible location of the insurance customer to be assigned. , Where, x c The x-coordinate or geographical longitude of the actual accessible location of the insurance customer to be assigned, y c f represents the vertical coordinates or geographical latitude and longitude of the actual accessible location of the insurance customer to be assigned. i Let x be the weight of the i-th first candidate position, n be the total number of first candidate positions, and x be the weight of the calculated position. i Let y be the x-coordinate or geographic longitude of the i-th first candidate location. i f is the ordinate or geographic latitude of the i-th first candidate location; where f i The basic weight f is calculated for the i-th first candidate position. i，基 With adjustable weight f i，调 The product of; f i，基 Let f be the ratio of the number of occurrences of the i-th first candidate calculated position to the total number of first candidate calculated positions. i，调 It is the product or sum of the periodic weighting coefficient, the dwell time weighting coefficient, and the frequency of occurrence weighting coefficient.

[0011] Optionally, the locations of all insurance clients visited by the salesperson within a preset second cycle from the last visit time are selected as the second candidate calculation locations, specifically including: Based on the time and location information of customer visits recorded by the application used by the salesperson to be assigned, construct the first time-series coordinate location set of the salesperson; In the first time-series coordinate location set of the salesperson, based on the visit time information of the customer visit, the second preset period is deduced backward from the last visit time of the salesperson to be assigned. According to the preset statistical period, the visit coordinate locations of all customers recorded by the application in each preset statistical period are selected to construct the second time-series coordinate location set of the salesperson in each preset statistical period. In each preset statistical period, select the coordinates of the salesperson's second time-series coordinates where the visit duration information is greater than a preset third duration threshold, and use them as the second candidate calculation positions in each preset statistical period.

[0012] Furthermore, selecting the visit locations of all insurance clients for the salesperson within a preset second cycle from the last visit time as the second candidate calculation locations specifically includes: If the salesperson is a new hire with no customer visit records, the second candidate calculation position is directly set to the coordinate position of the salesperson's marketing workplace, with the time sequence marked as the current month.

[0013] Optionally, the second candidate calculation position is calculated using a weighted calculation method based on the coordinate center point, to determine the specific center of the time-series activity area for the salesperson to be assigned: , Where, x e The x-coordinate or geographical longitude of the center of the activity area of ​​the salesperson to be assigned within each preset statistical period, y e The ordinate or geographic latitude of the center of the activity area of ​​the salesperson to be assigned within each preset statistical period, a j Let x be the weight of the j-th second candidate calculation position within each preset statistical period, m be the total number of second candidate calculation positions within each preset statistical period, and x be the weight of the j-th second candidate calculation position within each preset statistical period. j Let y be the x-coordinate or geographic longitude of the j-th second candidate calculation location within each preset statistical period. j The ordinate or geographic latitude of the j-th second candidate calculation location within each preset statistical period; a j The basic weight a of the j-th second candidate calculation position within each preset statistical period j，基 With adjustable weight a j，调 The product of; a j，基 a is the ratio of the number of occurrences of the j-th second candidate calculated position to the total number of second candidate calculated positions within each preset statistical period. j，调 It is the product or sum of the dwell time weighting coefficient and the frequency of occurrence weighting coefficient.

[0014] Optionally, the weighted distance parameters are calculated based on the actual accessible locations of the insurance customers to be assigned and the center of the activity area of ​​the agents to be assigned, specifically including: F = d1 × w1 + d2 × w2 d3×w3 d4×w4, Wherein, F is the total distance score, d1 is the straight-line distance in kilometers between the customer's actual accessible location and the center of the agent's time-series activity area in each preset statistical period, w1 is the weight of the straight-line distance in kilometers, d2 is the commuting distance in kilometers between the customer's actual accessible location and the center of the agent's time-series activity area in each preset statistical period, w2 is the weight of the commuting distance in kilometers, d3 is the first weighted coefficient for service convenience, w3 is the weight of the first weighted coefficient for service convenience, d4 is the second weighted coefficient for service convenience, w3 is the weight of the second weighted coefficient for service convenience; the first weighted coefficient for service convenience d3 is the number of other insurance customers served within a radius of a preset first kilometer, centered on the center of the agent's time-series activity area in each preset statistical period; the second weighted coefficient for service convenience d4 is the number of other insurance customers to be assigned within a radius of a preset first kilometer, centered on the actual accessible location of the insurance customer to be assigned.

[0015] Furthermore, based on the weighted distance parameters, the allocation of insurance customers to be assigned to sales agents specifically includes: Using the customer's actual accessible location as the center, within a radius of a preset first kilometer, starting from the current date and working backwards through the preset first period, query the activity area center of the salesperson with the highest cumulative frequency in all preset statistical periods during the period and the corresponding salesperson, and obtain the total distance score between the insurance customer to be assigned and the salesperson with the highest cumulative frequency to be assigned. If there are at least two sales representatives, the sales representative with the smallest total distance score is selected as the optimal follow-up service sales representative; if the total distance scores are the same, the average total distance score between the sales representative's activity area center and the customer to be assigned in each preset statistical period is calculated, and the sales representative with the smallest average total distance score is selected as the optimal follow-up service sales representative. If no salesperson is found, the search range will be gradually expanded based on the preset first kilometer radius.

[0016] A second aspect of the present invention provides a location-based computational insurance customer resource optimization allocation system, comprising: The acquisition module acquires the visit history information of insurance customers to be assigned and agents to be assigned. The historical information includes the historical information of visit location, visit time, and visit duration. The insurance customers to be assigned are policy customers whose original agents have left the company, or existing policy customers whose addresses have changed and who have not registered their new addresses with the insurance company. The first determination module selects the visit locations that appear more frequently than a preset first frequency threshold within a preset first period since the last visit time, based on the historical information of the visit locations and visit times of the insurance customers to be assigned. These locations are then used as the first candidate calculation locations. The first candidate calculation locations are calculated using a weighted calculation method based on the coordinate center point to determine the actual accessible locations of the insurance customers to be assigned. The second determination module selects the visit locations of all customers corresponding to the salesperson within a preset second period from the last visit time, based on the historical information of the visit location, visit time, and visit duration of the salesperson to be assigned. These locations are used as the second candidate calculation locations. The second candidate calculation locations are calculated using a weighted calculation method based on the coordinate center point. The center of the time sequence activity area of ​​the salesperson to be assigned is determined using the weighted calculation method based on the coordinate center point. The calculation and allocation module calculates distance parameters based on the actual accessible location of the insurance customers to be allocated and the center of the activity area of ​​the agents to be allocated in a time sequence; and allocates the insurance customers to the agents to be allocated based on the weighted distance parameters.

[0017] The technical solution adopted in this invention has the following technical effects: 1. In this invention, based on the historical information of the visit locations and visit times of the insurance customers to be assigned, visit locations with a frequency greater than a preset first frequency threshold within a preset first period from the last visit time are selected as first candidate calculation locations. These first candidate locations are then calculated using a weighted calculation method based on coordinate center points to determine the actual accessible locations of the insurance customers to be assigned. Based on the historical information of the visit locations and visit times of the salesperson to be assigned, visit locations of all insurance customers corresponding to that salesperson within a preset second period from the last visit time are selected as second candidate calculation locations. These second candidate locations are then calculated using a weighted calculation method based on coordinate center points to determine the temporal activity area center of the salesperson to be assigned. A weighted distance parameter is then calculated based on the actual accessible locations of the insurance customers to be assigned and the temporal activity area center of the salesperson to be assigned. Based on the weighted distance parameter, the insurance customers to be assigned are assigned to the salespersons to be assigned. This effectively solves the problem of low reliability in the optimal allocation of insurance customer resources caused by existing technologies, and effectively improves the reliability of the optimal allocation of insurance customer resources.

[0018] 2. The selection of the first candidate calculation position in the technical solution of this invention not only considers the frequency of occurrence and the effectiveness of visits, but also the actual situation such as the lack of any visit records for the insurance customers to be assigned. This not only improves the reliability of the optimized allocation of insurance customer resources, but also enhances its adaptability. Furthermore, by calculating the first candidate calculation position using the coordinate center point weighted calculation method, the weight of each first candidate calculation position in the calculation of the actual accessible position of the insurance customer to be assigned can be flexibly adjusted according to the frequency of occurrence, occurrence time, and dwell time weight coefficient of the first candidate calculation position, further improving the adaptability of the optimized allocation of insurance customer resources.

[0019] 3. The selection of the second candidate calculation position in the technical solution of this invention not only considers the visit duration, but also the actual situation such as the salesperson being a new employee. This not only improves the reliability of the optimized allocation of insurance customer resources, but also enhances its adaptability. Furthermore, the calculation of the second candidate calculation position using the coordinate center point weighted calculation method is performed according to a preset statistical period. The weight of each second candidate calculation position can be flexibly adjusted based on the frequency of occurrence of the second candidate calculation position, the dwell time weight coefficient, etc., further improving the adaptability of the optimized allocation of insurance customer resources.

[0020] 4. In this invention, the customer's actual accessible location is used as the center. Within a radius of a preset first kilometer, the system counts backwards from the current date to determine the activity center of the salesperson with the highest cumulative frequency across all preset statistical periods, along with the corresponding salesperson. The total distance score between the insurance customer to be assigned and the salesperson with the highest cumulative frequency is obtained. If at least two salespersons are found, the salesperson with the lowest total distance score is selected as the optimal follow-up service salesperson. If the total distance scores are the same, the average total distance score between the activity center of the salesperson's activity area and the customer to be assigned for each preset statistical period is calculated, and the salesperson with the lowest average total distance score is selected as the optimal follow-up service salesperson. If no salesperson is found, the query range is gradually expanded based on the preset first kilometer radius. This improves the adaptability of insurance customer resource optimization and allocation, reducing customer service costs, increasing customer service satisfaction, and improving customer re-development rates.

[0021] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating the method of Embodiment 1 in the present invention; Figure 2 This is a flowchart illustrating step S2 in the method of Embodiment 1 of the present invention; Figure 3 This is another flowchart illustrating step S2 in the method of Embodiment 1 of the present invention; Figure 4 This is a flowchart illustrating step S3 in the method of Embodiment 1 of the present invention; Figure 5 This is another flowchart illustrating step S3 in the method of Embodiment 1 of the present invention; Figure 6 This is a flowchart illustrating step S4 in the method of Embodiment 1 of the present invention; Figure 7 This is a schematic diagram of the system structure in Embodiment 2 of the present invention. Detailed Implementation

[0024] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. To simplify the disclosure of the invention, components and arrangements of specific examples are described below. Furthermore, reference numerals and / or letters may be repeated in different examples. This repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. Descriptions of well-known components, processing techniques, and processes are omitted in this invention to avoid unnecessarily limiting the invention.

[0025] Example 1 like Figure 1 As shown, this invention provides a location-based computational method for optimizing the allocation of insurance customer resources, comprising: S1, obtain the visit history information of insurance customers to be assigned and agents to be assigned; the historical information includes the historical information of visit location and visit time. The insurance customers to be assigned are policy customers whose original agents have left the company, or existing policy customers whose addresses have changed and who have not registered the new addresses with the insurance company. S2. Based on the historical information of the visit location and visit time of the insurance customers to be assigned, select the visit location that appears more frequently than the preset first frequency threshold within the preset first cycle since the last visit time as the first candidate calculation location. Calculate the first candidate calculation location using the coordinate center point weighted calculation method to determine the actual accessible location of the insurance customers to be assigned. S3. Based on the historical information of the visit location and visit time of the salesperson to be assigned, select the visit locations of all insurance customers corresponding to the salesperson within the preset second cycle from the last visit time as the second candidate calculation location. Calculate the second candidate calculation location using the coordinate center point weighted calculation method to determine the center of the time sequence activity area of ​​the salesperson to be assigned. S4. Based on the actual accessible location of the insurance customer to be assigned and the center of the activity area of ​​the agent to be assigned, calculate the distance parameters with weights; based on the weighted distance parameters, assign the insurance customer to the agent to be assigned.

[0026] In step S1, the visit history information of the insurance customers to be assigned and the sales agents to be assigned is obtained. The history information includes the historical information of the visit location, the historical information of the visit time, and the historical information of the visit duration. The insurance customers to be assigned are the policy customers whose original sales agents have left the company, or the existing policy customers whose addresses have changed and have not registered the new addresses with the insurance company. That is, when the policy customers of an agent leave the company and need to be reassigned to other sales agents, or when the addresses of existing customers change and are inconsistent with the addresses registered on the policy and have not registered their new addresses with the insurance company in a timely manner, this method can be used to reassign the relevant policy customers to the most suitable sales agents for service, thereby reducing customer service costs, improving customer service satisfaction, and increasing the customer re-development rate.

[0027] Over the years, life insurance agents have accumulated a wealth of customer location information through using the app developed by their life insurance company to visit clients. This information can be used to determine the client's accessible location and the agent's operational area using a method that calculates the most recently used coordinate center point. Therefore, when a policyholder needs to be reassigned to another agent, a location-based comprehensive algorithm can be used to assign the client to the most suitable agent for service.

[0028] Among them, such as Figure 2 As shown, in step S2, selecting the visit location whose frequency of occurrence within a preset first period from the last visit time is greater than a preset first frequency threshold as the first candidate calculation location specifically includes: S201, construct a set of the first coordinate locations of the customers based on the visit location information of all agents corresponding to the insurance customers to be assigned, recorded by the application; S202, In the first set of customer coordinate locations, based on the visit time information of the customer visit, reverse the first preset period from the visit time of the last insurance customer to be assigned, select all coordinate locations recorded by the application within the preset first period, and construct the second set of customer coordinate locations; Specifically, based on the system time of customer visits recorded by the insurance company app used by the salesperson, three quarters can be deduced from the last customer visit time (preset as the first cycle), and all coordinate locations recorded by the app during this period can be selected. The three-quarter limit is because life insurance companies generally stipulate that the frequency of customer visits should be at least once per quarter. Other time periods can also be specified, but this embodiment does not impose any restrictions. That is, the first candidate calculation location for calculating the customer's actual accessible location is selected based on the customer's "most recent" visit location.

[0029] S203, select coordinate locations from the customer's second set of coordinate locations whose frequency of occurrence (the number of times the coordinate location appears within three quarters prior to the last customer visit) is greater than a first preset percentage of the total number of visits to the customers to be assigned as insurance. These locations are then selected as the first candidate calculation locations. The total number of visits refers to visits confirmed by the customers to be assigned through on-site QR code scanning or visits whose duration exceeds a preset first duration threshold. In other words, the first candidate calculation location is selected based on the customer's "most frequently visited" location for calculating the customer's actual accessible location.

[0030] Specifically, the first preset percentage can be 30%, and the first duration threshold can be 10 minutes. These can also be flexibly adjusted according to the actual situation, and this embodiment does not impose any restrictions on them.

[0031] If multiple coordinate locations in the second set of customer coordinate locations appear more frequently than the first preset percentage of the total number of visits to the customer to be assigned, then all of them are included; the set of the first candidate calculation locations is as follows: P={(x1, y1), (x2, y2)...(x n y n )}.

[0032] Preferably, such as Figure 3 As shown, in step S2, selecting the visit location whose frequency of occurrence within a preset first period from the last visit time is greater than a preset first frequency threshold as the first candidate calculation location specifically includes: S204. If there is no coordinate location in the customer coordinate location set that has a frequency greater than the first preset percentage of the total number of visits to the insurance customers to be assigned within the first preset period, then select the preset first number (e.g., 3) of coordinate locations in the customer coordinate location set that are closest to the last visit time in chronological order. S205 If there are no visit records for the insurance customer to be assigned, the address registered on the most recent policy of the insurance customer to be assigned shall be converted into location coordinates and used as the first candidate location for calculation.

[0033] Preferably, if the first candidate calculated location is an outlier (significantly deviating from the city range), it is excluded. Specifically, if a first candidate calculated location is filtered by the latitude and longitude range of the city boundary and is more than 10 kilometers away from the city boundary of the customer to be assigned, it is considered an outlier and is discarded. Alternatively, a single deviation threshold of 10% can be used: if the deviation of a first candidate calculated location from the average distance of all candidate calculated locations exceeds 10%, it is considered an outlier and excluded.

[0034] To reduce computational complexity, when the number of first candidate computation positions exceeds the preset second number (e.g., 3), the top 3 can be included in the weighted calculation in descending order of frequency (if the frequency difference between the 3rd and 4th positions is ≤5%, then all of them are included), to avoid excessive computational complexity. The preset second number can be flexibly adjusted according to the actual amount of data and computation.

[0035] In step S2, the first candidate calculation location is calculated using a weighted calculation method based on the coordinate center point, and the actual contactable location of the insurance customer to be assigned is determined as follows: , Where, x c The x-coordinate or geographical longitude of the actual accessible location of the insurance customer to be assigned, y c f represents the vertical coordinates or geographical latitude and longitude of the actual accessible location of the insurance customer to be assigned. i Let x be the weight of the i-th first candidate position, n be the total number of first candidate positions, and x be the weight of the calculated position. i Let y be the x-coordinate or geographic longitude of the i-th first candidate location. i f is the ordinate or geographic latitude of the i-th first candidate location; where f i The basic weight f is calculated for the i-th first candidate position. i，基 With adjustable weight f i，调 The product of; f i，基 Let f be the ratio of the number of occurrences of the i-th first candidate calculated position to the total number of first candidate calculated positions. i，调 It is the product or sum of the periodic weighting coefficient, the dwell time weighting coefficient, and the frequency of occurrence weighting coefficient.

[0036] Specifically, based on the weighted calculation method of the coordinate center point, the coordinates of the customer's actual accessible location are calculated according to the customer's "most recent" and "most frequently visited" locations.

[0037] The periodic weighting coefficient is as follows: when the time corresponding to the first candidate calculation position and the time since the last visit are less than 1 / 2 of the preset first time period, the periodic weighting coefficient is the sum of 1 and the preset second percentage (e.g., 20%); when the time corresponding to the first candidate calculation position and the time since the last visit are not less than 1 / 2 of the preset first time period, the periodic weighting coefficient is the difference between 1 and the preset second percentage. The dwell time weighting coefficient is as follows: when the dwell time corresponding to the first candidate calculation location is less than the preset second dwell time threshold (e.g., 15 minutes, which is greater than the preset first dwell time threshold), the dwell time weighting coefficient is the sum of 1 and the preset third percentage (e.g., 20%); when the dwell time corresponding to the first candidate calculation location is not less than the preset second dwell time threshold, the periodic weighting coefficient is the difference between 1 and the preset third percentage. The frequency weighting coefficient is as follows: when the frequency of occurrence corresponding to the first candidate calculation position is less than the preset second frequency threshold (greater than the preset first frequency threshold, for example, greater than 40% of the total number of visits to insurance customers to be assigned), the frequency weighting coefficient is the sum of 1 and the preset fourth percentage (for example, 20%); when the frequency of occurrence corresponding to the first candidate calculation position is not less than the preset second frequency threshold, the frequency weighting coefficient is the difference between 1 and the preset fourth percentage.

[0038] The preset second percentage, preset third percentage, and preset fourth percentage can be the same or different, and can be flexibly adjusted according to the actual situation. This embodiment does not impose any restrictions on them.

[0039] Among them, such as Figure 4 As shown, in step S3, the locations of all insurance clients visited by the salesperson within a preset second period from the last visit time are selected as the second candidate calculation locations, specifically including: S301, Based on the time and location information of customer visits recorded by the salespersons to be assigned using the application, construct the first time-series coordinate location set of the salespersons; S302, In the first time-series coordinate position set of salespersons, based on the visit time information of customer visits, the second preset period is deduced backward from the last visit time of the salesperson to be assigned, and the visit coordinate positions of all customers recorded by the application in each preset statistical period are selected according to the preset statistical period, and the second time-series coordinate position set of salespersons in each preset statistical period is constructed. Based on the system time of customer visits recorded by the app used by the salesperson, count back three quarters from the last customer visit time (preset second cycle), select all coordinate positions recorded by the app in each month of the period (preset statistical cycle is month), and form the monthly activity "time-series coordinate set" of the salesperson for the coordinates of all customer visit positions that fall into the set of coordinates of the salesperson. That is, the second time-series coordinate position set of the salesperson. For example, if a salesperson has customer visit coordinates (x1, y1) and (x2, y2) in month n1 of year m1, then the "second time-series coordinate position set" of the salesperson in month n1 of year m1 is {(m1, n1): (x1, y1), (x2, y2)}.

[0040] The salesperson's "time-series coordinates" can be stored in a unified format: "timestamp (yyyyMMddHH:mm:ss) + latitude and longitude (accurate to 6 decimal places)". Application scenario: Used to analyze salespersons' "weekly activity patterns on specific days" (e.g., concentrated visits to region A every Wednesday), matching high-frequency customer visit dates (e.g., customers are frequently at home on Wednesdays), improving allocation adaptability. The calculation rule for salespersons can be: counting backwards from the current date, with the 1st of each month as the node, and the preset statistical period can be a calendar month (e.g., March 1st to March 31st).

[0041] S303: Select the coordinates from the second time-series coordinates of the salesperson in each preset statistical period where the visit duration information is greater than the preset third duration threshold (e.g., 5 minutes) as the second candidate calculation position in each preset statistical period.

[0042] Furthermore, such as Figure 5 As shown, in step S3, selecting the visit locations of all insurance customers corresponding to the salesperson within a preset second cycle from the last visit time as the second candidate calculation location specifically includes: S304. If the salesperson is a new employee and has no customer visit records, the second candidate calculation position is directly set to the coordinate position of the salesperson's marketing workplace, and the time sequence is marked as the current month.

[0043] In step S3, the second candidate calculation position is calculated using a weighted calculation method based on the coordinate center point, and the specific center of the time-series activity area for the salesperson to be assigned is determined as follows: , Where, x e The x-coordinate or geographical longitude of the center of the activity area of ​​the salesperson to be assigned within each preset statistical period, y e The ordinate or geographic latitude of the center of the activity area of ​​the salesperson to be assigned within each preset statistical period, a j Let x be the weight of the j-th second candidate calculation position within each preset statistical period, m be the total number of second candidate calculation positions within each preset statistical period, and x be the weight of the j-th second candidate calculation position within each preset statistical period. j Let y be the x-coordinate or geographic longitude of the j-th second candidate calculation location within each preset statistical period. j The ordinate or geographic latitude of the j-th second candidate calculation location within each preset statistical period; a j The basic weight a of the j-th second candidate calculation position within each preset statistical period j，基 With adjustable weight a j，调 The product of; a j，基 a is the ratio of the number of occurrences of the j-th second candidate calculated position to the total number of second candidate calculated positions within each preset statistical period. j，调It is the product or sum of the dwell time weighting coefficient and the frequency of occurrence weighting coefficient.

[0044] The dwell time weighting coefficient is as follows: when the dwell time corresponding to the second candidate calculation location is less than the preset third dwell time threshold (e.g., 15 minutes), the dwell time weighting coefficient is the sum of 1 and the preset fifth percentage (e.g., 20%); when the dwell time corresponding to the second candidate calculation location is not less than the preset third dwell time threshold, the periodic weighting coefficient is the difference between 1 and the preset fifth percentage. The frequency weighting coefficient is as follows: when the frequency of occurrence corresponding to the second candidate calculation position is less than the preset third frequency threshold (which can be greater than or equal to or less than the preset first frequency threshold, for example, greater than 20% or 30% of the total number of visits by the salesperson to be assigned), the frequency weighting coefficient is the sum of 1 and the preset sixth percentage (for example, 20%); when the frequency of occurrence corresponding to the second candidate calculation position is not less than the preset third frequency threshold, the frequency weighting coefficient is the difference between 1 and the preset sixth percentage.

[0045] In step S4, the weighted calculation of distance parameters based on the actual accessible locations of the insurance customers to be assigned and the center of the activity area of ​​the agents to be assigned includes: F = d1 × w1 + d2 × w2 d3×w3 d4×w4, Wherein, F is the total distance score, d1 is the straight-line distance in kilometers between the customer's actual accessible location and the center of the agent's time-series activity area in each preset statistical period, w1 is the weight of the straight-line distance in kilometers, d2 is the commuting distance in kilometers between the customer's actual accessible location and the center of the agent's time-series activity area in each preset statistical period, w2 is the weight of the commuting distance in kilometers, d3 is the first weighted coefficient of service convenience, w3 is the weight of the first weighted coefficient of service convenience, d4 is the second weighted coefficient of service convenience, w3 is the weight of the second weighted coefficient of service convenience; the first weighted coefficient of service convenience d3 is the number of other insurance customers served within a radius of a preset first kilometer (e.g., 1 kilometer) centered on the center of the agent's time-series activity area in each preset statistical period to be assigned; the second weighted coefficient of service convenience d4 is the number of other insurance customers to be assigned within a radius of a preset first kilometer centered on the actual accessible location of the insurance customer to be assigned.

[0046] The preset first kilometer is a dynamically adjusted value, with the urban core area reduced to 0.5 kilometers, the general urban area to 1 kilometer, and the suburbs to 2 kilometers.

[0047] In the formula, the closer the distance and the more customers served along the route, the lower the score and the higher the priority for allocation. The straight-line distance can be calculated using the Haversine formula for latitude and longitude, while the commuting distance is obtained by calling the map API to get the shortest distance. The weight values ​​(w1, w2, w3, w4, the sum of which is 1) can be determined using historical customer visit data based on regression analysis algorithms. Because salespeople have diverse travel methods and flexibly choose transportation tools based on actual distance, salespeople can manually select their mode of transport (driving / cycling / walking) through the app, or the system can infer it based on "coordinate movement speed" (e.g., speed > 30 km / h is considered driving). Weight adjustment: When driving, the commuting distance weight w2 can be reduced to 0.3, and the straight-line distance weight w1 can be increased to 0.5; when walking, w1 is reduced to 0.2, and w2 is increased to 0.6. Other adjustment rules can also be used, for example... ,in, Maximum setting for straight-line distance in kilometers; , Maximum setting for commuting distance in kilometers; , The maximum number of insurance customers that can be served can be defined within a radius of a preset first kilometer (e.g., 1 kilometer) centered on the customer's actual contact location. , The maximum number of insurance customers to be assigned is defined by taking the actual accessible location of the customers to be assigned as the center and a preset first kilometer as the radius; the specific adjustment rules are not limited in this embodiment of the invention.

[0048] Among them, such as Figure 6 As shown, in step S4, assigning insurance customers to agents based on the weighted distance parameters specifically includes: S401: Using the customer's actual accessible location as the center, within a radius of a preset first kilometer, starting from the current date and working backwards through the preset first period, query the activity area center of the salesperson with the highest cumulative frequency in all preset statistical periods during the period and the corresponding salesperson, and obtain the total distance score between the insurance customer to be assigned and the salesperson with the highest cumulative frequency to be assigned. Using the coordinates of the customer's physically accessible location as the center, within a radius of 1 kilometer (preset initial kilometer), counting backwards three quarters from the current date, the system retrieves the center coordinates of the activity areas of the most frequently occurring salespersons for all months within that period, along with the associated salespersons. It also calculates the total distance score between the assigned insurance customer and that salesperson for each month. To maintain the affiliation between the customer and the insurance company, the salesperson is limited to those within the customer's sales workplace. The radius is adjusted in increments of 0.5 kilometers (e.g., 1 km → 1.5 km → 2 km…). The maximum radius is 5 kilometers. If there are still fewer than 3 salespersons, cross-workplace assignments are made (prioritizing workplaces adjacent to the customer's).

[0049] S402, if there are at least two salespersons, the salesperson with the smallest total distance score is selected as the optimal follow-up service salesperson; if the total distance scores are the same, the average total distance score between the salesperson's activity area center and the customer to be assigned in each preset statistical period is calculated, and the salesperson with the smallest average total distance score is selected as the optimal follow-up service salesperson. When at least two agents are present, allocation can be based on the matching degree between customer and agent profile tags. For example, customer tags can be defined as follows: "Customer Star Rating" (based on a score of 50% of premiums + 30% of renewal rate + 20% of referrals, divided into 5 levels), "Policy Type" (life insurance / health insurance / accident insurance, etc.), and "Age Group" (18-25 years old = 1, 26-35 years old = 2, ..., 60 years and above = 5). Agent tags can be defined as follows: "Customer Occupation" (the occupation that appears most frequently among customers served in the past year), "Specialized Insurance Type" (the insurance type with the highest sales performance), and "Service Star Rating" (based on customer satisfaction score). In other words, when at least two agents are present, allocation can be based on tag priority. For example, allocation can be prioritized by age group, policy type, interests, etc., or by agent's current customer load rate (lower load preferred), and historical satisfaction rating of similar customers (≥4.5 points preferred). If the scores are still the same, random allocation is performed and the reason is recorded (for subsequent model optimization).

[0050] S403 If no salesperson is found, the search range will be gradually expanded based on the preset first kilometer radius.

[0051] Maximum radius: no more than 5 kilometers; if there are still fewer than 3 salespersons, they will be assigned across workplaces (prioritizing workplaces adjacent to the client's workplace).

[0052] It should be noted that the location coordinate data sourced from the app in this invention requires the following accuracy: GPS error ≤ 10 meters, WiFi positioning error ≤ 50 meters. Data cleaning: Automatically remove duplicate coordinates (with identical timestamps and latitude / longitude) and drifting coordinates with weak signals (three consecutive positioning deviations > 500 meters). In special scenarios where customer coordinate information is missing (unauthorized positioning): the city / region and policy address corresponding to the IP address of the most recent premium payment are used for determination.

[0053] In this invention, based on the historical information of the visit locations and visit times of the insurance customers to be assigned, visit locations that appear more frequently than a preset first frequency threshold within a preset first period since the last visit are selected as first candidate calculation locations. These first candidate locations are then calculated using a weighted calculation method based on coordinate center points to determine the actual accessible locations of the insurance customers to be assigned. Next, based on the historical information of the visit locations and visit times of the salesperson to be assigned, visit locations of all insurance customers corresponding to that salesperson within a preset second period since the last visit are selected as second candidate calculation locations. These second candidate locations are then calculated using a weighted calculation method based on coordinate center points to determine the temporal activity area center of the salesperson to be assigned. Finally, a weighted distance parameter is calculated based on the actual accessible locations of the insurance customers to be assigned and the temporal activity area center of the salesperson to be assigned. Based on the weighted distance parameter, the insurance customers to be assigned are assigned to the salespersons to be assigned. This effectively solves the problem of low reliability in the optimal allocation of insurance customer resources caused by existing technologies, and effectively improves the reliability of the optimal allocation of insurance customer resources.

[0054] The selection of the first candidate calculation position in the technical solution of this invention not only considers the frequency of occurrence and the effectiveness of visits, but also the actual situation such as the lack of any visit records for the insurance customers to be assigned. This not only improves the reliability of the optimal allocation of insurance customer resources, but also enhances its adaptability. Furthermore, by calculating the first candidate calculation position using the coordinate center point weighted calculation method, the weight of each first candidate calculation position in the calculation of the actual accessible position of the insurance customer to be assigned can be flexibly adjusted according to the frequency of occurrence, occurrence time, dwell time weight coefficient, etc. of the first candidate calculation position, further improving the adaptability of the optimal allocation of insurance customer resources.

[0055] The selection of the second candidate calculation position in the technical solution of this invention not only considers the visit duration but also the actual situation such as the salesperson being a new employee. This not only improves the reliability of the optimized allocation of insurance customer resources but also enhances its adaptability. Furthermore, the calculation of the second candidate calculation position using the coordinate center point weighted calculation method is performed separately according to a preset statistical period. The weight of each second candidate calculation position can be flexibly adjusted based on factors such as the frequency of occurrence of the second candidate calculation position and the weight coefficient of the dwell time, further improving the adaptability of the optimized allocation of insurance customer resources.

[0056] In this invention, the technical solution uses the customer's actual accessible location as the center. Within a radius of a preset first kilometer, starting from the current date and working backwards for a preset first period, it queries the activity area center of the salesperson with the highest cumulative frequency across all preset statistical periods during that period, along with the corresponding salesperson. It also obtains the total distance score between the insurance customer to be assigned and the salesperson with the highest cumulative frequency. If at least two salespersons are found, the salesperson with the lowest total distance score is selected as the optimal follow-up service salesperson. If the total distance scores are the same, the average total distance score between the activity area center of the salesperson and the customer to be assigned for each preset statistical period is calculated, and the salesperson with the lowest average total distance score is selected as the optimal follow-up service salesperson. If no salesperson is found, the query range is gradually expanded based on the preset first kilometer radius. This improves the adaptability of insurance customer resource optimization and allocation, reducing customer service costs, increasing customer service satisfaction, and improving customer re-engagement rates.

[0057] Example 2 like Figure 7 As shown, the present invention also provides a location-based insurance customer resource optimization and allocation system, comprising: The acquisition module 101 acquires the visit history information of insurance customers to be assigned and agents to be assigned; the historical information includes visit location history information, visit time history information and visit duration history information. The insurance customers to be assigned are policy customers whose original agents have left the company, or existing policy customers whose addresses have changed and who have not registered their new addresses with the insurance company. The first determining module 102 selects, based on the historical information of the visit location and the historical information of the visit time of the insurance customer to be assigned, the visit location that appears more frequently than a preset first frequency threshold within a preset first period since the last visit time as the first candidate calculation location. The first candidate calculation location is calculated using a weighted calculation method based on the coordinate center point to determine the actual accessible location of the insurance customer to be assigned. The second determining module 103 selects the visit locations of all customers corresponding to the salesperson within a preset second period from the last visit time, based on the historical information of the visit location, the historical information of the visit time, and the historical information of the visit duration of the salesperson to be assigned, as the second candidate calculation location. The second candidate calculation location is calculated using the coordinate center point weighted calculation method, and the center of the time sequence activity area of ​​the salesperson to be assigned is determined using the coordinate center point weighted calculation method. The calculation and allocation module 104 calculates distance parameters based on the actual accessible location of the insurance customer to be allocated and the center of the activity area of ​​the agent to be allocated in the time sequence; and allocates the insurance customer to the agent to be allocated based on the weighted distance parameters.

[0058] The implementation process of the acquisition module 101, the first determination module 102, the second determination module 103, and the calculation and allocation module 104 in this embodiment 2 corresponds to the method steps in embodiment 1, and will not be repeated here.

[0059] In this invention, based on the historical information of the visit locations and visit times of the insurance customers to be assigned, visit locations that appear more frequently than a preset first frequency threshold within a preset first period since the last visit are selected as first candidate calculation locations. These first candidate locations are then calculated using a weighted calculation method based on coordinate center points to determine the actual accessible locations of the insurance customers to be assigned. Next, based on the historical information of the visit locations and visit times of the salesperson to be assigned, visit locations of all insurance customers corresponding to that salesperson within a preset second period since the last visit are selected as second candidate calculation locations. These second candidate locations are then calculated using a weighted calculation method based on coordinate center points to determine the temporal activity area center of the salesperson to be assigned. Finally, a weighted distance parameter is calculated based on the actual accessible locations of the insurance customers to be assigned and the temporal activity area center of the salesperson to be assigned. Based on the weighted distance parameter, the insurance customers to be assigned are assigned to the salespersons to be assigned. This effectively solves the problem of low reliability in the optimal allocation of insurance customer resources caused by existing technologies, and effectively improves the reliability of the optimal allocation of insurance customer resources.

[0060] The selection of the first candidate calculation position in the technical solution of this invention not only considers the frequency of occurrence and the effectiveness of visits, but also the actual situation such as the lack of any visit records for the insurance customers to be assigned. This not only improves the reliability of the optimal allocation of insurance customer resources, but also enhances its adaptability. Furthermore, by calculating the first candidate calculation position using the coordinate center point weighted calculation method, the weight of each first candidate calculation position in the calculation of the actual accessible position of the insurance customer to be assigned can be flexibly adjusted according to the frequency of occurrence, occurrence time, dwell time weight coefficient, etc. of the first candidate calculation position, further improving the adaptability of the optimal allocation of insurance customer resources.

[0061] The selection of the second candidate calculation position in the technical solution of this invention not only considers the visit duration but also the actual situation such as the salesperson being a new employee. This not only improves the reliability of the optimized allocation of insurance customer resources but also enhances its adaptability. Furthermore, the calculation of the second candidate calculation position using the coordinate center point weighted calculation method is performed separately according to a preset statistical period. The weight of each second candidate calculation position can be flexibly adjusted based on factors such as the frequency of occurrence of the second candidate calculation position and the weight coefficient of the dwell time, further improving the adaptability of the optimized allocation of insurance customer resources.

[0062] In this invention, the technical solution uses the customer's actual accessible location as the center. Within a radius of a preset first kilometer, starting from the current date and working backwards for a preset first period, it queries the activity area center of the salesperson with the highest cumulative frequency across all preset statistical periods during that period, along with the corresponding salesperson. It also obtains the total distance score between the insurance customer to be assigned and the salesperson with the highest cumulative frequency. If at least two salespersons are found, the salesperson with the lowest total distance score is selected as the optimal follow-up service salesperson. If the total distance scores are the same, the average total distance score between the activity area center of the salesperson and the customer to be assigned for each preset statistical period is calculated, and the salesperson with the lowest average total distance score is selected as the optimal follow-up service salesperson. If no salesperson is found, the query range is gradually expanded based on the preset first kilometer radius. This improves the adaptability of insurance customer resource optimization and allocation, reducing customer service costs, increasing customer service satisfaction, and improving customer re-engagement rates.

[0063] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for optimizing the allocation of insurance customer resources based on location computing, characterized in that, include: Obtain visit history information of insurance customers to be assigned and agents to be assigned; the historical information includes visit location history information and visit time history information. Insurance customers to be assigned are policy customers whose original agents have left the company, or existing policy customers whose addresses have changed and who have not registered the new addresses with the insurance company. Based on the historical information of the visit location and visit time of the insurance customers to be assigned, the visit location that appears more frequently than a preset first frequency threshold within a preset first cycle since the last visit time is selected as the first candidate calculation location. The first candidate calculation location is calculated using the coordinate center point weighted calculation method to determine the actual accessible location of the insurance customers to be assigned. Based on the historical information of the visit locations and visit times of the salesperson to be assigned, the visit locations of all insurance customers of the salesperson within the second cycle from the last visit time are selected as the second candidate calculation locations. The second candidate calculation locations are calculated using the coordinate center point weighted calculation method to determine the center of the time sequence activity area of ​​the salesperson to be assigned. Based on the actual accessible location of the insurance customers to be assigned and the center of the activity area of ​​the agents to be assigned, a weighted distance parameter is calculated; based on the weighted distance parameter, the insurance customers to be assigned are assigned to the agents to be assigned.

2. The method for optimizing the allocation of insurance customer resources based on location computing as described in claim 1, characterized in that, The locations where the frequency of a visit exceeds a preset first frequency threshold within a preset first period from the last visit time are selected as the first candidate calculation locations. Specifically, these include: Based on the visit location information of all agents corresponding to the insurance customers to be assigned, using the application to record customer visits, construct a set of the customer's first coordinate location; In the first set of customer coordinate locations, based on the visit time information of the customer visit, the first preset period is deduced by going back from the visit time of the last insurance customer to be assigned, and all coordinate locations recorded by the application within the first preset period are selected to construct the second set of customer coordinate locations. Select the coordinates in the second set of customer coordinates that appear more frequently than a first preset percentage of the total number of visits to the insurance customers to be assigned, and use them as the first candidate calculation locations; where the total number of visits is the number of visits confirmed by the insurance customers to be assigned on-site by scanning the code or the visit duration information is greater than a preset first duration threshold.

3. The method for optimizing the allocation of insurance customer resources based on location computing according to claim 2, characterized in that, The selection of visit locations that have appeared more frequently than a preset first frequency threshold within a preset first period since the last visit, as the first candidate calculation locations, specifically includes: If there is no coordinate location in the customer coordinate location set that has a frequency greater than the first preset percentage of the total number of visits to the insurance customers to be assigned within the first preset period, then select the preset first number of coordinate locations in the customer coordinate location set that are closest to the last visit time in chronological order. If there are no visit records for the insurance customer to be assigned, the address registered on the customer's most recent policy will be converted into location coordinates and used as the first candidate location for calculation.

4. The method for optimizing the allocation of insurance customer resources based on location calculation according to claim 2, characterized in that, The first candidate calculation location is calculated using a weighted calculation method based on the coordinate center point, and the actual contactable location of the insurance customer to be assigned is determined as follows: ， Where, x c The x-coordinate or geographical longitude of the actual accessible location of the insurance customer to be assigned, y c f represents the vertical coordinates or geographical latitude and longitude of the actual accessible location of the insurance customer to be assigned. i Let x be the weight of the i-th first candidate position, n be the total number of first candidate positions, and x be the weight of the calculated position. i Let y be the x-coordinate or geographic longitude of the i-th first candidate location. i f is the ordinate or geographic latitude of the i-th first candidate location; where f i The basic weight f is calculated for the i-th first candidate position. i，基 With adjustable weight f i，调 The product of; f i，基 Let f be the ratio of the number of occurrences of the i-th first candidate calculated position to the total number of first candidate calculated positions. i，调 It is the product or sum of the periodic weighting coefficient, the dwell time weighting coefficient, and the frequency of occurrence weighting coefficient.

5. The method for optimizing the allocation of insurance customer resources based on location computing according to claim 1, characterized in that, The locations of all insurance clients visited by the salesperson within a pre-defined second period since the last visit are selected as the second candidate calculation locations, specifically including: Based on the time and location information of customer visits recorded by the application used by the salesperson to be assigned, construct the first time-series coordinate location set of the salesperson; In the first time-series coordinate location set of the salesperson, based on the visit time information of the customer visit, the second preset period is deduced backward from the last visit time of the salesperson to be assigned. According to the preset statistical period, the visit coordinate locations of all customers recorded by the application in each preset statistical period are selected to construct the second time-series coordinate location set of the salesperson in each preset statistical period. In each preset statistical period, select the coordinates of the salesperson's second time-series coordinates where the visit duration information is greater than a preset third duration threshold, and use them as the second candidate calculation positions in each preset statistical period.

6. The method for optimizing the allocation of insurance customer resources based on location computing according to claim 5, characterized in that, The second candidate calculation location is selected from all insurance clients visited by the salesperson within a preset second period since the last visit. This includes: If the salesperson is a new hire with no customer visit records, the second candidate calculation position is directly set to the coordinate position of the salesperson's marketing workplace, with the time sequence marked as the current month.

7. The method for optimizing the allocation of insurance customer resources based on location computing according to claim 5, characterized in that, The second candidate calculation position is calculated using a weighted calculation method based on the coordinate center point, and the specific center of the time sequence activity area for the salesperson to be assigned is determined as follows: ， Where, x e The x-coordinate or geographical longitude of the center of the activity area of ​​the salesperson to be assigned within each preset statistical period, y e The ordinate or geographic latitude of the center of the activity area of ​​the salesperson to be assigned within each preset statistical period, a j Let x be the weight of the j-th second candidate calculation position within each preset statistical period, m be the total number of second candidate calculation positions within each preset statistical period, and x be the weight of the j-th second candidate calculation position within each preset statistical period. j Let y be the x-coordinate or geographic longitude of the j-th second candidate calculation location within each preset statistical period. j The ordinate or geographic latitude of the j-th second candidate calculation location within each preset statistical period; a j The basic weight a of the j-th second candidate calculation position within each preset statistical period j，基 With adjustable weight a j，调 The product of; a j，基 a is the ratio of the number of occurrences of the j-th second candidate calculated position to the total number of second candidate calculated positions within each preset statistical period. j，调 It is the product or sum of the dwell time weighting coefficient and the frequency of occurrence weighting coefficient.

8. The method for optimizing the allocation of insurance customer resources based on location computing according to claim 5, characterized in that, Based on the actual accessible locations of the insurance customers to be assigned and the center of the activity area of ​​the agents to be assigned, the weighted distance parameters specifically include: F=d1×w1+d2×w2 d3×w3 d4×w4, Wherein, F is the total distance score, d1 is the straight-line distance in kilometers between the customer's actual accessible location and the center of the agent's time-series activity area in each preset statistical period, w1 is the weight of the straight-line distance in kilometers, d2 is the commuting distance in kilometers between the customer's actual accessible location and the center of the agent's time-series activity area in each preset statistical period, w2 is the weight of the commuting distance in kilometers, d3 is the first weighted coefficient for service convenience, w3 is the weight of the first weighted coefficient for service convenience, d4 is the second weighted coefficient for service convenience, w3 is the weight of the second weighted coefficient for service convenience; the first weighted coefficient for service convenience d3 is the number of other insurance customers served within a radius of a preset first kilometer, centered on the center of the agent's time-series activity area in each preset statistical period; the second weighted coefficient for service convenience d4 is the number of other insurance customers to be assigned within a radius of a preset first kilometer, centered on the actual accessible location of the insurance customer to be assigned.

9. A method for optimizing the allocation of insurance customer resources based on location computing as described in claim 8, characterized in that, Based on the weighted distance parameters, the insurance customers to be assigned are specifically assigned to the sales agents to be assigned, including: Using the customer's actual accessible location as the center, within a radius of a preset first kilometer, starting from the current date and working backwards through the preset first period, query the activity area center of the salesperson with the highest cumulative frequency in all preset statistical periods during the period and the corresponding salesperson, and obtain the total distance score between the insurance customer to be assigned and the salesperson with the highest cumulative frequency to be assigned. If there are at least two sales representatives, the sales representative with the smallest total distance score is selected as the optimal follow-up service sales representative; if the total distance scores are the same, the average total distance score between the sales representative's activity area center and the customer to be assigned in each preset statistical period is calculated, and the sales representative with the smallest average total distance score is selected as the optimal follow-up service sales representative. If no salesperson is found, the search range will be gradually expanded based on the preset first kilometer radius.

10. A location-based computing-based insurance customer resource optimization allocation system, characterized in that, include: The acquisition module acquires the visit history information of insurance customers to be assigned and agents to be assigned. The historical information includes the historical information of visit location, visit time, and visit duration. The insurance customers to be assigned are policy customers whose original agents have left the company, or existing policy customers whose addresses have changed and who have not registered their new addresses with the insurance company. The first determination module selects the visit locations that appear more frequently than a preset first frequency threshold within a preset first period since the last visit time, based on the historical information of the visit locations and visit times of the insurance customers to be assigned. These locations are then used as the first candidate calculation locations. The first candidate calculation locations are calculated using a weighted calculation method based on the coordinate center point to determine the actual accessible locations of the insurance customers to be assigned. The second determination module selects the visit locations of all customers corresponding to the salesperson within a preset second period from the last visit time, based on the historical information of the visit location, visit time, and visit duration of the salesperson to be assigned. These locations are used as the second candidate calculation locations. The second candidate calculation locations are calculated using a weighted calculation method based on the coordinate center point. The center of the time sequence activity area of ​​the salesperson to be assigned is determined using the weighted calculation method based on the coordinate center point. The calculation and allocation module calculates distance parameters based on the actual accessible location of the insurance customers to be allocated and the center of the activity area of ​​the agents to be allocated in a time sequence; and allocates the insurance customers to the agents to be allocated based on the weighted distance parameters.

Images