A big data-based customer management method and system

Abstract

This invention discloses a customer management method and system based on big data, relating to the field of customer management technology. The method includes: assigning values ​​to the opening association identifiers of seats based on the opening contour data of the CAD drawing in the target venue and the Wi-Fi access point of each seat; determining the structural exposure weight of the Wi-Fi access point to the opening based on the opening association identifiers; writing the opening transmission table based on the opening association identifiers and structural exposure weights; calculating the first cumulative dwell time of anonymous devices based on the session logs of the target venue and the structural exposure weights in the opening transmission table, and calculating the second cumulative dwell time. This invention improves the accuracy of identifying the actual customer dwell area.

Description

Technical Field

[0001] This invention relates to the field of customer management technology, and in particular to a customer management method and system based on big data. Background Technology

[0002] In large stadiums, commercial complexes, and cultural and sports facilities, customer activity areas typically encompass a complex space comprised of grandstands, corridors, and various access openings. The complexity of these spatial structures directly impacts customer behavior identification, location assessment, and resource allocation. With the increasing intelligence of venues, Wi-Fi networks have become a crucial infrastructure for user behavior perception. Numerous terminal devices connect wirelessly to perform positioning, navigation, and interactive operations, generating massive amounts of connection data based on session logs. Against this backdrop, precise management of customer behavior through big data analytics has become a vital means of improving venue operational efficiency and optimizing services. However, traditional methods primarily rely on the geographical location of access points and connection time to predict customer behavior, making it difficult to identify the interference effects of low-obstruction channels within the building structure on signal propagation paths and customer dwell time assessments. This results in insufficient accuracy and lag in customer zoning, resource allocation, and service provision.

[0003] Current customer management technologies generally lack a modeling mechanism for the relationship between signal propagation paths in complex spatial structures and the actual location of customers. This leads to situations where, when customers connect to the network through Wi-Fi access points, some signals may be misattributed to areas other than their actual location due to the penetration effect caused by open structures, resulting in customer location identification errors. In this case, resource allocation and operational decisions based on the access point's geographical location are severely biased, impacting operational efficiency. Existing methods struggle to quantify the physical connection strength between Wi-Fi access points and the channel structure, and fail to decompose and reconstruct customer dwell behavior according to structural paths, thus limiting the implementation of precise service delivery and dynamic resource allocation strategies. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing customer management technologies, which generally lack a modeling mechanism for the relationship between signal propagation paths and actual customer locations in complex structural spaces. Therefore, this invention proposes a customer management method and system based on big data.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution:

[0006] A big data-based customer management method includes:

[0007] S1. Assign values ​​to the opening association identifiers of seats based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat;

[0008] S2. Determine the structural exposure weight of the Wi-Fi access point to the opening based on the opening association identifier;

[0009] S3. Write the opening transfer table according to the opening association identifier and the structure exposure weight;

[0010] S4. Calculate the first cumulative dwell time of the anonymous device based on the session log and the structure exposure weight in the open transmission table of the target venue, and calculate the second cumulative dwell time;

[0011] S5. Calculate the resource quota for each opening based on the first cumulative stay time and the total resource quantity of the target venue;

[0012] S6. Based on the second cumulative stay duration and resource quota, perform operational resource matching and service delivery to customer zones of the target venue.

[0013] Preferably, the opening association identifier of each seat is assigned a value based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat, including:

[0014] Polygon extraction is performed on the opening contour data of the CAD drawing to obtain multiple openings;

[0015] Extract the edge line from the boundary of the opening;

[0016] Starting from the direction of the stands, perform a morphological dilation operation on the edge line based on the row spacing of the seats in the target stadium to obtain the shaded area;

[0017] Determine the spatial overlap between the Wi-Fi access point and the shaded area for each seat;

[0018] The opening association identifier of the seat is assigned a value based on the judgment result of the spatial overlap relationship.

[0019] Preferably, the opening association identifier of the seat is assigned a value based on the judgment result of the spatial overlap relationship, including:

[0020] If the determination result is that the location coordinates of the Wi-Fi access point fall within the shaded area, then the seat opening association identifier is set to 1; otherwise, the seat opening association identifier is set to 0.

[0021] Preferably, the structural exposure weight of the Wi-Fi access point to the opening is determined based on the opening association identifier, including:

[0022] Define the surrounding corridor area of ​​the target venue;

[0023] Discrete ray emission is performed at each Wi-Fi access point;

[0024] After the discrete ray emission is completed, the number of rays that pass through the opening and reach the corridor area is counted according to the opening association marker;

[0025] The number of rays is normalized to obtain the structural exposure weight of the Wi-Fi access point for the opening.

[0026] Preferably, the opening transfer table is written according to the opening association identifier and the structure exposure weight, including:

[0027] If the Wi-Fi access point is associated with the opening of the seat as 1, then the structural exposure weight of the Wi-Fi access point to the opening is written into the corresponding cell of the opening transmission table.

[0028] If the Wi-Fi access point's association identifier with the seat opening is 0, then write 0 to the corresponding cell in the opening transmission table.

[0029] Preferably, the first cumulative dwell time of the anonymous device is calculated based on the session logs and structural exposure weights in the open transmission table of the target venue, including:

[0030] Obtain the session logs of the target venue, which include the anonymous device ID, Wi-Fi access point number, connection start time, and connection end time.

[0031] The difference between the connection end time and the connection start time is calculated to obtain the dwell time of the anonymous device at the Wi-Fi access point;

[0032] The dwell time and the structural exposure weight in the aperture transfer table are multiplied to obtain the allocated time of the anonymity device in each aperture direction;

[0033] The allocated time is accumulated based on the opening number of the opening to obtain the first cumulative dwell time of the anonymous device.

[0034] Preferably, calculating the second cumulative stay duration includes:

[0035] The structural exposure weights in the aperture transfer table are summed to obtain the total aperture transfer weights;

[0036] Subtract the sum of the open transmission weights from 1 to get the remaining proportion of the grandstand area;

[0037] The time allotted for the grandstand area is obtained by multiplying the dwell time and the remaining proportion of the grandstand area.

[0038] The duration allocated to the grandstand area is accumulated based on the anonymous device ID to obtain the second cumulative dwell time of the anonymous device.

[0039] Preferably, the resource quota for each opening is calculated based on the first cumulative stay duration and the total resource availability of the target venue, including:

[0040] The operating intensity of each opening is obtained by summing the first cumulative dwell time of all anonymous devices;

[0041] The total number of openings in the entire site is obtained by summing the operating intensity of all openings.

[0042] The ratio of operating intensity to the total number of openings in the entire site is calculated to obtain the relative share of each opening;

[0043] Obtain the total resources of the target venue;

[0044] The resource allocation value is obtained by multiplying the total resource amount and the relative share of each opening.

[0045] Round the resource allocation value down to get the resource quota for each opening.

[0046] Preferably, operational resources are matched and services are deployed to customer zones within the target venue based on the second cumulative stay duration and resource quota, including:

[0047] Sum the second cumulative dwell time of all anonymous devices to obtain the total dwell time in the stands;

[0048] Based on the total time spent in the stands, customer service needs are prioritized and divided into high-service zones, medium-service zones, and low-service zones.

[0049] Develop differentiated operational strategies for high-service, medium-service, and low-service zones;

[0050] Resources are allocated to the high-service, medium-service, and low-service partitions based on the resource quota for each opening.

[0051] To address the above problems, the present invention also provides a customer management system based on big data, the system comprising:

[0052] The opening association recognition module is used to assign values ​​to the opening association identifier of each seat based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat.

[0053] The structure exposure calculation module is used to determine the structure exposure weight of the Wi-Fi access point to the opening based on the opening association identifier;

[0054] The transmission mapping generation module is used to write the opening transmission table according to the opening association identifier and the structure exposure weight;

[0055] The location duration accounting module is used to calculate the first cumulative dwell time of the anonymous device based on the session log of the target venue and the structure exposure weight in the opening transmission table, and to calculate the second cumulative dwell time.

[0056] The resource share calculation module is used to calculate the resource quota for each opening based on the first cumulative stay time and the total resource amount of the target venue;

[0057] The operations matching and delivery module is used to match operational resources and deliver services to customer segments of the target venue based on the second cumulative stay time and resource quota.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] 1. In this invention, by introducing opening contour data, edge lines, and morphological expansion algorithms from CAD drawings, a physical mapping mechanism between the venue's spatial structure and the Wi-Fi signal propagation path is established. This effectively identifies the impact of building openings on signal penetration behavior. By judging spatial overlap relationships, an opening association identifier is assigned to the Wi-Fi access point of each seat. Furthermore, the number of discrete rays penetrating the opening to the corridor area is counted, realizing the quantification of structural exposure weight. This compensates for the lack of spatial structural constraint modeling in traditional customer identification methods that rely solely on the geographical location of access points, improves the accuracy of identifying the actual customer residence area, avoids cross-area misattribution caused by signal penetration misjudgment, and fundamentally solves the problem of insufficient identification accuracy in the background technology.

[0060] 2. In this invention, by constructing an opening transmission table with structural exposure weight as its core, and combining it with the connection time information of anonymous devices in the session log, this invention proposes a dwell time calculation method based on structural propagation capability, namely, a method for constructing the first cumulative dwell time and the second cumulative dwell time. This method of integrating physical structural propagation characteristics into dwell behavior calculation realizes the projection and reconstruction of customer behavior from the access point level to the structural opening dimension, forming a distribution map reflecting the intensity of customer spatial behavior. It can quantify the operational intensity of each opening, accurately calculate the resource quota of the structural space, and thereby realize the partitioned management of customer service needs and the on-demand allocation of operational resources, thus improving resource utilization efficiency and customer service response capabilities, and effectively addressing the problem of poor resource allocation accuracy in existing technologies.

[0061] 3. In this invention, a service demand priority division mechanism based on the total dwell time of the entire stadium is further constructed based on the second cumulative dwell time. The target venue is divided into high-service zones, medium-service zones, and low-service zones. Combined with the number of open-air resource slots, differentiated operation strategies and resource allocation plans are formulated. By investing more guidance resources and marketing methods in high-density dwell areas, and implementing energy-saving or guidance operation methods in low-activity areas, dynamic allocation of resources and guidance and control of customer behavior are realized. This achieves refined stratification of customer service and efficient allocation of operational resources, thereby improving venue operation efficiency and customer satisfaction. Attached Figure Description

[0062] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0063] Figure 1 This is a flowchart illustrating a customer management method based on big data, provided as an embodiment of the present invention.

[0064] Figure 2 This is a functional module diagram of a big data-based customer management system provided in one embodiment of the present invention. Detailed Implementation

[0065] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0066] Example: This example provides a customer management method based on big data. See [link / reference] Figure 1 Specifically, including:

[0067] S1. Assign values ​​to the opening association identifiers of seats based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat;

[0068] In an embodiment of the present invention, the opening association identifier of each seat is assigned a value based on the opening contour data of the CAD drawing in the target venue and the Wi-Fi access point of each seat, including:

[0069] Polygon extraction is performed on the opening contour data of the CAD drawing to obtain multiple openings;

[0070] Specifically, the process begins by acquiring the CAD drawing of the target venue. This CAD drawing stores the opening outline data, which records the geometric boundary information of the passageways between the stands and the surrounding corridors. Then, computer-aided design software or a program with graphics processing capabilities is used to analyze the opening outline data of the CAD drawing, identifying the set of geometric lines that constitute the boundaries of each opening. Next, a polygon construction algorithm is used to connect these geometric lines sequentially into closed polygon structures. Each closed polygon corresponds to one opening, resulting in multiple openings. This clarifies the shape and location of each opening in the actual space of the target venue, providing precise spatial units for subsequent analysis of the impact of the openings on wireless signal propagation and customer location attribution.

[0071] Specifically, the CAD drawing of the target venue refers to the architectural structural data file of the venue recorded in the computer-aided design format. This file contains the geometric shape, spatial dimensions, and structural hierarchy information of each functional area inside the venue, and is the basic data source for venue spatial modeling and structural analysis. The opening refers to the passage space formed between the stands and the corridor as described in the CAD drawing. It is enclosed by the structural boundary and has the function of personnel passage. It is usually used for the entry and exit and flow of the audience between the stands and the corridor. Geometrically, the opening is a passage area formed by the interruption of the building wall. It has low obstruction characteristics in the process of wireless signal propagation and is an important spatial factor leading to cross-zone signal leakage.

[0072] Specifically, the opening outline data in the CAD drawing refers to the geometric boundary information of the passage between the stands and the corridor inside the target stadium, stored in the form of electronic architectural design files. This data reflects the shape and location of the opening in the actual space. Polygon extraction refers to extracting a closed polygon structure that can fully express the spatial range of the opening from this geometric boundary information, which is used to form the spatial unit of the opening.

[0073] Extract the edge line from the boundary of the opening;

[0074] Specifically, after obtaining the closed polygonal structure with multiple openings, the boundary data of each opening polygon is first acquired. This data contains all the geometric line segments that constitute the boundary of the opening. Then, the geometric feature analysis algorithm is used to identify these boundary line segments one by one. Based on the architectural structural features at the intersection of the stands and the corridor in the target stadium, the line segments located at the intersection of the stands and the corridor are selected. The linear geometric elements composed of these line segments are the edge lines, which accurately define the actual boundary range of the opening and provide a reference path for subsequent morphological operations and the generation of the shadow zone area.

[0075] Specifically, the rim line refers to the linear geometric element formed around the boundary of the grandstand opening in the architectural structure of the target venue. This line is located at the intersection between the grandstand and the surrounding corridor and is used to accurately define the actual boundary range of the opening. From a spatial perspective, the rim line represents the structural outline of the opening and is the dividing line between the audience area and the passageway area. In geometric modeling, it is used to determine the starting boundary of the signal propagation direction and the reference path for morphological operations. Its position directly affects the subsequent generation of the shadow zone area and the identification of the low-obstacle passageway range.

[0076] Starting from the direction of the stands, perform a morphological dilation operation on the edge line based on the row spacing of the seats in the target stadium to obtain the shaded area;

[0077] Specifically, the row spacing of seats starting from the stands refers to the center-to-center distance formed by the rows of seats in the target stadium. This distance reflects the density and hierarchy of the audience seating arrangement. Morphological dilation calculation refers to the geometric calculation process of generating an extended area within a fixed distance along the edge of the opening towards the stands. This is used to simulate the reduced occlusion space caused by the opening. The shadow zone area refers to the strip-shaped area obtained through dilation calculation. This area corresponds to the low-obstruction passage range created by the lack of spectators and seats on the stands. It is a key geometric entity for identifying signal propagation leakage paths.

[0078] Specifically, first, the direction of the target stadium's stands is determined, i.e., the direction facing the seating arrangement in the stands; then, the row spacing of the seats in the target stadium is obtained, which is the center-to-center distance formed by the rows of seats; then, using morphological operation tools, starting from the extracted edge line, an expansion operation is performed along the stand direction according to the obtained row spacing of the seats, so that the edge line expands along the stand direction to an area corresponding to the row spacing of the seats; finally, the strip-shaped area obtained through this expansion operation is the shadow area, which corresponds to the low-obstruction passage range in the stands where there are no seats or human bodies to block the way due to the opening, and is used to identify the key geometric entities of the wireless signal propagation leakage path.

[0079] Determine the spatial overlap between the Wi-Fi access point and the shaded area for each seat;

[0080] Specifically, the actual installation coordinates of the Wi-Fi access points for each seat are first obtained from the equipment deployment files of the target venue. Simultaneously, the geometric boundary data of each shadow zone region, generated earlier through morphological dilation calculations, is retrieved. This geometric boundary data includes the vertex coordinates and boundary range parameters of the closed polygons constituting the shadow zone region. Then, the Wi-Fi access point coordinates of the seats and the geometric boundary data of the shadow zone regions are unified to the same Cartesian coordinate system to ensure consistent spatial positioning and avoid judgment errors caused by differences in coordinate systems. Next, for each seat's Wi-Fi access point, a Boolean geometric intersection operation is used to calculate whether there is a spatial intersection between the access point coordinates and the range enclosed by the geometric boundaries of each shadow zone region. If the coordinates of the access point fall within the geometric boundary of a certain shaded area, or if the coordinates of the access point partially overlap with the boundary of the shaded area, then it is determined that the Wi-Fi access point of the seat has a spatial overlap relationship with the shaded area. If the coordinates of the access point are completely outside the geometric boundary of all shaded areas, then it is determined that there is no spatial overlap relationship. Finally, the spatial overlap judgment results of the Wi-Fi access points of each seat are recorded, and the access points with spatial overlap relationships are associated with the opening identifiers of the corresponding shaded areas. At the same time, an AP-opening overlap relationship table is generated, which clearly records the overlap status of the Wi-Fi access point of each seat with each shaded area, providing data support for subsequently assigning opening association identifiers to the seats.

[0081] Specifically, the Wi-Fi access point for the seats refers to the wireless network access device installed under or near the audience seats in the target venue. This device is used to provide local wireless coverage for the audience in the stands, and its location corresponds one-to-one with the seat arrangement. Spatial overlap relationship judgment refers to determining whether the access point is within the low-obstacle passage influence range formed by the opening by calculating whether there is an intersection between the coordinate position of the access point and the geometric range of the shadow zone in digital spatial modeling. Seat opening association identification refers to the logical identification information assigned to each access point or its corresponding seat after the spatial overlap relationship judgment is completed. This is used to identify whether the access point is spatially associated with a specific opening. The identification result will serve as the basis data for subsequent signal propagation modeling and structural exposure weight calculation.

[0082] The opening association identifier of the seat is assigned a value based on the judgment result of the spatial overlap relationship.

[0083] In an embodiment of the present invention, assigning a value to the seat opening association identifier based on the judgment result of the spatial overlap relationship includes:

[0084] If the determination result is that the location coordinates of the Wi-Fi access point fall within the shaded area, then the seat opening association identifier is set to 1; otherwise, the seat opening association identifier is set to 0.

[0085] Specifically, the shaded area refers to the low-obstruction passageway formed by openings in the target venue. Wireless signal propagation is weakly obstructed within this area. If the Wi-Fi access point coordinates of a seat fall within this area, there is a risk of signal leakage to the surrounding corridor area through the opening. This could lead to anonymous devices connected to that access point being incorrectly identified as residing in the stands, distorting customer location data. If the location does not fall within this area, the signal is strongly obstructed by seats and people, resulting in a lower risk of leakage and relatively accurate customer location data. Therefore, by assigning an opening-related identifier of 1 to Wi-Fi access points of seats falling within the shaded area and 0 to those not falling within, access points with signal leakage risk can be clearly identified. This provides basic identifier data for subsequent calculations of structural exposure weights, construction of opening transmission tables, and correction of spatial assignment of customer dwell time. This enables precise management of customer locations, solves the problem of misclassification of customer zones due to signal leakage, and ensures the accuracy and reliability of customer location data in big data-based customer management methods.

[0086] Specifically, assigning opening association values ​​to seats based on the opening contour data from the CAD drawings of the target venue and the Wi-Fi access points of each seat is to accurately identify which seat Wi-Fi access points are within the low-obstruction passageway influence range of the grandstand openings at the spatial structure level. Since the signal propagation characteristics of seat access points rely on the occupant's body and the seat's occupancy effect, when the opening area breaks this occupancy continuity, access points near the opening will generate a stronger direct-line propagation path in the opening direction, and their signals can easily penetrate into the corridor area and form cross-zone connections with equipment within the corridor. By analyzing the opening contour data in the CAD drawings and combining it with the spatial coordinates of the access points, assigning opening association values ​​to the access points corresponding to each seat allows for the establishment of a correspondence between structural geometry and signal propagation at the data level. This provides the foundation for subsequent structural exposure weight calculations, opening transmission table generation, and cross-zone connection correction, achieving accurate modeling of the low-obstruction passageway effect and improving the accuracy of customer location identification.

[0087] S2. Determine the structural exposure weight of the Wi-Fi access point to the opening based on the opening association identifier;

[0088] In an embodiment of the present invention, determining the structural exposure weight of the Wi-Fi access point to the opening based on the opening association identifier includes:

[0089] Define the surrounding corridor area of ​​the target venue;

[0090] Discrete ray emission is performed at each Wi-Fi access point;

[0091] Specifically, the ring corridor area of ​​the target venue refers to the long, narrow passageway area located around the stands, connecting multiple entrances and exits, and used for the flow and dispersal of crowds. This area is usually connected to the stands openings and is the main passageway for spectators to enter and exit the seating area. Its continuous and unobstructed structure makes it easy to become a concentrated coverage and propagation path for wireless signals. Discrete ray emission for each Wi-Fi access point refers to using the access point as the signal source and generating multiple straight paths at equal intervals in the horizontal direction from its location to simulate the radiation and propagation process of the signal. This process is used to evaluate whether the signal can pass through the stands openings and reach the ring corridor area, thereby analyzing the access point's accessibility and exposure level in this area. This is an important computational basis for subsequently establishing a quantitative model of structural exposure weight and spatial penetration capability.

[0092] Specifically, defining the ring corridor area of ​​the target venue involves analyzing the structural layer data in the venue's CAD drawing, extracting the boundary of the passageway area connecting the outer perimeter of the stands with multiple openings, and marking this area as a public passageway for pedestrian dispersion according to the venue's functional zoning standards. The geometric range of this area will be used as the signal propagation endpoint area in subsequent judgments. Discrete ray emission for each Wi-Fi access point refers to constructing several equally angled rays from the spatial coordinates of the access point, spaced at intervals from 0 degrees to 360 degrees. Each ray simulates the propagation path of the wireless signal in that direction. The system then determines whether each ray passes through the opening structure boundary and extends into the predefined ring corridor area without interruption. The number of rays entering is counted as a signal reachability index, providing basic data for subsequent structural exposure weight calculations.

[0093] After the discrete ray emission is completed, the number of rays that pass through the opening and reach the corridor area is counted according to the opening association marker;

[0094] Specifically, after the discrete ray emission ends, the system first filters out the set of opening targets to be included in the statistics based on the opening association identifier between each Wi-Fi access point and the structural opening. Then, it performs path tracking on each ray originating from the access point to determine whether it continuously crosses the boundary line of the corresponding opening and extends into the predefined corridor area without interruption. In this judgment process, the structural position through which the ray passes must be within the opening area and the endpoint coordinates must fall within the corridor area. Rays that meet the conditions are counted as a valid penetration event. The system accumulates the count of all rays that meet the above conditions and finally obtains the number of rays that each access point penetrates each opening and reaches the corridor area. This statistical result serves as the basic data input for subsequent structural exposure weight calculation.

[0095] The number of rays is normalized to obtain the structural exposure weight of the Wi-Fi access point for the opening.

[0096] Specifically, the number of rays passing through the opening and reaching the corridor area refers to the number of discrete rays originating from each Wi-Fi access point that successfully cross the boundary of the structural opening and ultimately extend into the corridor area. This number reflects the potential ability of the access point's signal to penetrate into the public area through a specific structural channel. Ray count normalization refers to standardizing the statistical count of all rays to express the relative accessibility of each access point to different openings on a uniform numerical scale. The structural exposure weight of the Wi-Fi access point to the opening refers to the weight obtained after the above normalization, which is used to measure the degree to which a specific access point's signal is exposed to the corridor area through a certain opening in the structural space. The higher the weight, the easier it is for the access point to penetrate the opening, which may affect the customer management system's judgment of the device's belonging area.

[0097] Specifically, normalizing the number of rays involves, based on obtaining the number of rays penetrating all structural openings for each Wi-Fi access point, first calculating the sum of all effective rays corresponding to that access point. Then, using the number of rays penetrating each opening as the numerator and the total number of effective rays for that access point as the denominator, a proportional calculation is performed to obtain the structural exposure ratio value of each opening relative to that access point. This ratio value is the structural exposure weight, representing the relative intensity of the access point's signal penetrating a specific opening and being exposed to the outside. This weight value is limited to the range of zero to one, reflecting the degree of contribution of each opening to the signal propagation path of the access point in the structural space. It is an important basis for subsequently constructing the transmission mapping relationship and calculating the positioning deviation correction.

[0098] Specifically, determining the structural exposure weight of Wi-Fi access points to openings based on opening association markers aims to establish a quantifiable mapping between spatial structure and wireless propagation, thereby identifying the actual impact of low-resistance paths created by openings on signal penetration. Since Wi-Fi access points in the lower seating areas of large venues are spatially adjacent to multiple openings, their signals are highly likely to penetrate structural openings and leak into nearby public areas or even cross-zone connections, thus interfering with the judgment of the audience's actual location. Opening association markers clarify the spatial connection relationships between access points and each opening. Based on this, the number of paths through which signals penetrate openings and enter the surrounding corridor area can be counted and normalized to form structural exposure weights. This reflects the contribution of each access point to signal penetration in the structural space for each opening, providing data support for subsequent location correction, cross-zone connection identification, and improved customer management accuracy.

[0099] S3. Write the opening transfer table according to the opening association identifier and the structure exposure weight;

[0100] In an embodiment of the present invention, the writing process of the opening transfer table according to the opening association identifier and the structure exposure weight includes:

[0101] If the Wi-Fi access point is associated with the opening of the seat as 1, then the structural exposure weight of the Wi-Fi access point to the opening is written into the corresponding cell of the opening transmission table.

[0102] If the Wi-Fi access point's association identifier with the seat opening is 0, then write 0 to the corresponding cell in the opening transmission table.

[0103] Specifically, the structural exposure weight of the Wi-Fi access point's association with the seat opening is written into the corresponding cell of the opening transmission table. This ensures that the signal transmission mapping strength is established only when a clear structural channel connection exists, thus avoiding invalid or erroneous associations that could interfere with positioning accuracy. When there is a physical line-of-sight path between the access point and the opening, and the signal is indeed likely to penetrate the opening and propagate to the surrounding corridor area, the system writes the structural contribution of this path into the transmission table in the form of a normalized exposure weight to express the signal penetration capability of the path. When there is no direct structural association between the access point and the opening, i.e., the opening association is zero, the system writes a zero value into the transmission table, indicating that there is no effective signal reachability for this path and it should not be included in the propagation model. This approach ensures that the information recorded in the transmission table has structural authenticity and logical propagation rationality, providing an important data foundation for subsequent session displacement correction and customer management analysis.

[0104] Specifically, writing the opening transmission table based on the opening association identifier and structural exposure weight is to construct a fixed mapping matrix that represents both spatial structural constraints and signal penetration capabilities. This matrix accurately reflects the connection relationship and propagation strength between each Wi-Fi access point and the surrounding corridor area through structural openings. The opening association identifier determines whether an access point has spatial reachability through a specific opening, while the structural exposure weight quantifies the signal exposure degree of that propagation path at the structural level. Combining these two elements and writing them into the transmission table allows the system to infer user spatial location and reassign session behavior directly based on structural information during subsequent processing, without relying on traditional positioning algorithms. This transmission table, as the sole physical mapping basis for session migration and customer partitioning, significantly improves the stability, interpretability, and accuracy of customer management methods in complex structural environments, and is a key technical means to achieve cross-regional connection correction and resource allocation optimization.

[0105] S4. Calculate the first cumulative dwell time of the anonymous device based on the session log and the structure exposure weight in the open transmission table of the target venue, and calculate the second cumulative dwell time;

[0106] In an embodiment of the present invention, calculating the first cumulative dwell time of the anonymous device based on the session log of the target venue and the structural exposure weight in the open transmission table includes:

[0107] Obtain the session logs of the target venue, which include the anonymous device ID, Wi-Fi access point number, connection start time, and connection end time.

[0108] Specifically, to achieve refined collection and analysis of customer behavior within the target venue, the first step is to acquire session log data from the venue using a big data acquisition system. This process involves collecting wireless connection behavior of customer devices through Wi-Fi access points deployed throughout the venue and uploading connection events to a unified data storage platform in real-time or near real-time. Each session log entry contains an anonymous device identifier automatically generated and anonymized by the system to uniquely identify the terminal device and prevent user privacy leaks. It also records the Wi-Fi access point number the device is connected to, which can be mapped to the venue's geospatial deployment information for further spatial location association. Furthermore, it records the start and end times of the connection event, stored as high-precision timestamps to ensure the accuracy and continuity of subsequent calculations. These logs can be centrally collected through a network monitoring module, the Wi-Fi controller log export interface, or the real-time streaming channel of the big data platform. After data cleaning and archiving, the data enters the data processing flow for customer management and analysis, forming a structured, large-scale connection behavior database. This step provides a complete data foundation for subsequently building a structure-aware customer flow model and precise resource allocation strategies.

[0109] Specifically, the target venue's session log refers to the set of raw connection data automatically recorded by the network access system when a user device connects to a Wi-Fi access point within the venue. The anonymous device ID is used to uniquely identify the terminal device but does not contain user identity information, ensuring privacy and security. The Wi-Fi access point number indicates the location of the wireless infrastructure node the user is currently connected to. The connection start time and connection end time correspond to the precise timestamps of the device connecting to and disconnecting from the node, respectively, and the time difference between the two reflects the duration the device spends in a particular area. This log serves as a crucial data source for constructing customer behavior spatiotemporal trajectories, analyzing structural correlation paths, and classifying customers for management purposes. It features continuity, objectivity, and high resolution, forming the foundation for customer location and operational resource scheduling under structural constraints.

[0110] The difference between the connection end time and the connection start time is calculated to obtain the dwell time of the anonymous device at the Wi-Fi access point;

[0111] The dwell time and the structural exposure weight in the aperture transfer table are multiplied to obtain the allocated time of the anonymity device in each aperture direction;

[0112] Specifically, connection end time and connection start time refer to the timestamp data recorded when an anonymous device connects to and subsequently disconnects from a Wi-Fi access point at a certain moment in the venue. The actual dwell time of the device at that access point can be calculated by subtracting these two timestamps. This dwell time reflects the intensity of the device's presence at a specific spatial location. Structure exposure weight is a normalized weight representing the ability of a Wi-Fi access point to propagate to a target area through structural openings, used to characterize the strength of structured connections. By multiplying the device's dwell time at the access point with the structure exposure weight corresponding to each opening, the dwell behavior can be mapped to each opening direction according to the structured connection ratio, thus obtaining the allocated duration of the device in all opening directions. This allocated duration constitutes a fine-grained breakdown of the device's structured dwell behavior, serving as the foundational data for subsequent customer path reconstruction and zoning management.

[0113] The allocated time is accumulated based on the opening number of the opening to obtain the first cumulative dwell time of the anonymous device.

[0114] Specifically, the opening number is a unique coded label used to identify each spatial opening in the venue structure leading to the corridor or internal passageway. Each number corresponds to an actual structural opening. By summing up the allocated time corresponding to all structural opening numbers, the total dwell time associated with that opening by multiple anonymous devices within a certain time window can be calculated. This sum is defined as the first cumulative dwell time, which reflects the crowd-attracting capacity of the structural opening within a certain period. It is an important data foundation for further evaluating customer behavior distribution and prioritizing operational resource allocation.

[0115] Specifically, the calculation of the first cumulative dwell time for anonymous devices is based on the principle of linear superposition in the physical propagation path and the statistical law of energy distribution superposition. Since the signal propagation capabilities of Wi-Fi access points differ in the direction of each structural opening in the venue, and the dwell time of each anonymous device at the access point can be considered as the total dwell time within the coverage area of ​​that access point, when considering how this dwell time spreads through the building structure, the dwell time should be proportionally allocated to all relevant opening numbers according to the structural exposure weight of each opening direction. This weight essentially reflects the geometric reachability and path transparency of signal propagation through the opening, possessing the significance of propagation efficiency on a physically reachable path. The time share allocated to each opening number is accumulated along the opening number dimension, equivalent to superimposing the sum of the dwell times of multiple anonymous devices in each opening direction. The final value reflects the actual number of customers converging in that structural opening direction, and is therefore defined as the first cumulative dwell time, possessing the ability to represent the dwell time distribution on the physical path.

[0116] Specifically, calculating the first cumulative dwell time of anonymous devices based on the target venue's session logs and the structural exposure weights in the opening transmission table is to further project connection behavior, which was originally limited to the access point level, onto specific structural opening directions, thereby accurately characterizing the dwell intensity of individual customers in different physical spatial paths. Session logs provide the connection times of devices at various Wi-Fi access points, serving as the fundamental data source for customer behavior. The structural exposure weights in the opening transmission table reflect the physical accessibility mapping relationship between different openings in the building structure and connection behavior. Combining these two methods allows for the indirect reconstruction of customer spatial flow trajectories, forming a dwell intensity distribution along various opening directions. This provides crucial data support for subsequent spatial heat identification, access bottleneck analysis, key access area identification, and precise delivery of venue zoned services. This approach overcomes the problems of coarse granularity and vague coverage in traditional AP-level customer analysis, achieving refined customer management under structural awareness, which is beneficial for improving venue operational efficiency and service quality.

[0117] In an embodiment of the present invention, calculating the second cumulative stay duration includes:

[0118] The structural exposure weights in the aperture transfer table are summed to obtain the total aperture transfer weights;

[0119] Subtract the sum of the open transmission weights from 1 to get the remaining proportion of the grandstand area;

[0120] Specifically, the structural exposure weight in the opening transmission table refers to the spatial accessibility measure between a Wi-Fi access point and each structural opening. A higher value indicates a higher probability that the access point can connect to the surrounding corridor area through the structural opening. This value is obtained through ray emission statistical normalization. The total opening transmission weight refers to the sum of the structural exposure weights of a given access point across all structural openings, reflecting the overall accessibility of the access point to communicate externally through each opening. The remaining proportion of the grandstand area obtained by subtracting the total opening transmission weight represents the percentage of the access point's ability to transmit its signal externally without any structural opening. This proportion can be used to measure whether the activities of the terminal devices served by the access point are limited to the internal area; a higher value indicates that the device user is more likely to be in structurally obstructed areas such as the inside of the grandstand. Quantitative analysis of these indicators can be used to further identify hidden user behavior patterns and for regional operation scheduling and strategy setting in customer management.

[0121] The time allotted for the grandstand area is obtained by multiplying the dwell time and the remaining proportion of the grandstand area.

[0122] The duration allocated to the grandstand area is accumulated based on the anonymous device ID to obtain the second cumulative dwell time of the anonymous device.

[0123] Specifically, multiplying the dwell time by the remaining proportion of the grandstand area yields the estimated dwell time of the device within the grandstand area, known as the grandstand area allocated dwell time. Summing up all such allocated dwell times based on the anonymous device identifier provides the device's second cumulative dwell time within the grandstand area throughout the entire session. This is used to assess whether the device's behavioral trajectory is concentrated in the internal enclosed area, thereby supporting data insights and behavioral profiling modeling for customer management.

[0124] Specifically, to estimate the dwell behavior of anonymous devices in the target stadium's grandstand area, indirect measurement based on structural characteristics is required. First, the structural exposure weights in the opening transmission table are summed to obtain the total transmission capacity value between the Wi-Fi access points and all openings within the stadium, i.e., the total weight. This value reflects the total probability that different access points will propagate signals through openings to the outside of the stadium or passageways. Then, the maximum transmission capacity value under ideal, fully exposed conditions is subtracted from the sum of the transmission weights to obtain the proportion of the access point that failed to complete signal transmission under structural obstruction, i.e., the remaining proportion in the grandstand area. This proportion indicates the degree to which the signal failed to transmit from the access point and remained in the enclosed grandstand area; a higher value indicates more severe obstruction. The Wi-Fi connection dwell time is then multiplied by this proportion to obtain the estimated dwell time of the device in the enclosed area. Finally, all corresponding dwell times are accumulated according to device identification to form a second accumulated dwell time, used to quantify customer activity in areas not covered by openings, thus providing implicit data support for customer management regarding obstructed spaces.

[0125] S5. Calculate the resource quota for each opening based on the first cumulative stay time and the total resource quantity of the target venue;

[0126] In an embodiment of the present invention, the resource quota for each opening is calculated based on the first cumulative stay duration and the total resource quantity of the target venue, including:

[0127] The operating intensity of each opening is obtained by summing the first cumulative dwell time of all anonymous devices;

[0128] The total number of openings in the entire site is obtained by summing the operating intensity of all openings.

[0129] The ratio of operating intensity to the total number of openings in the entire site is calculated to obtain the relative share of each opening;

[0130] Specifically, the first cumulative dwell time of anonymous devices refers to the total dwell time of each anonymous mobile terminal within the target venue in the spatial area surrounding its connected Wi-Fi access point, reflecting the cumulative physical activity of the device in the actual structural environment. The sum of this dwell time across all devices constitutes the total cumulative foot traffic within the areas covered by different openings, serving as the operational intensity of each opening. Operational intensity represents the integrated effect of the opening area's attractiveness to customers; a higher intensity value indicates that the opening is more frequently approached or occupied by customers. The total number of openings in the entire venue refers to the sum of the operational intensities of all openings, used to quantify the total customer agglomeration effect generated by the entire venue under structural guidance. The ratio between operational intensity and total number of openings in the entire venue is the relative share of each opening, reflecting the proportion of customer operations for each opening in the overall environment, and is a fundamental parameter for the rational allocation of subsequent operational resources. This share, as a standardized indicator, ensures that resources are accurately allocated according to the actual distribution of customer traffic.

[0131] Obtain the total resources of the target venue;

[0132] The resource allocation value is obtained by multiplying the total resource amount and the relative share of each opening.

[0133] Round the resource allocation value down to get the resource quota for each opening.

[0134] Specifically, total resources refer to the total number of limited operational resources available to the operator within the target venue and allocated to various service nodes. These resources can be digital screen slots, the number of guides, or other specific resource units. The resource allocation value is a floating-point number obtained by multiplying the total resources by the relative share of each opening, representing the amount of resources each opening should receive based on its ability to attract customers. The resource slots are the integer result after rounding down the resource allocation value, ensuring the feasibility and completeness of resource allocation in actual physical application. This parameter reflects the quantitative allocation of resources based on structural and behavioral data analysis and serves as the basic unit for subsequent customer management, operational execution, and venue scheduling.

[0135] Specifically, the resource quota for each opening is calculated by multiplying its relative contribution to the overall operating system by the total configurable resources and then rounding down. This method aligns with the principle of maximizing benefit equilibrium in resource allocation. In physical systems, resource allocation often employs a proportional allocation model. This involves setting an overall resource pool as a total limit, normalizing the performance factors of each unit (such as customer dwell time or structural guidance capabilities) to form a weighted distribution, and then determining the theoretical resource requirement for each unit by multiplying the weights by the total resources. Since actual resource units are typically discrete integers, rounding down is necessary to ensure the feasibility of the allocation. This method not only meets the requirements of fairness and efficiency maximization but also provides a closed-loop feedback mechanism using customer behavior data and structural attribute data. This offers a foundation for refined control over customer guidance, advertising placement, and personnel scheduling in venue operations, enhancing the scientific and precise nature of customer management.

[0136] S6. Based on the second cumulative stay duration and resource quota, perform operational resource matching and service delivery to customer zones of the target venue.

[0137] In an embodiment of the present invention, operational resources are matched and services are deployed to customer zones of the target venue based on the second cumulative stay duration and resource quota, including:

[0138] Sum the second cumulative dwell time of all anonymous devices to obtain the total dwell time in the stands;

[0139] Based on the total time spent in the stands, customer service needs are prioritized and divided into high-service zones, medium-service zones, and low-service zones.

[0140] Specifically, the process of prioritizing service demands by customer zones based on the total dwell time in the stands is achieved through statistical sorting and threshold stratification of customer dwell behavior data in each area of ​​the target venue. First, the system normalizes the second cumulative dwell time of all openings or areas to obtain the relative dwell percentage of each zone, representing the density of customer activity in different areas. Then, the system sorts all areas from high to low according to the dwell percentage and determines stratification thresholds based on statistical distribution characteristics. For example, it can divide the overall data into three intervals based on the cumulative distribution curve of the dwell percentage or the natural discontinuity method. Areas in the upper interval are defined as high-service zones, indicating high customer concentration, frequent activity, and high resource demand; areas in the middle interval are defined as medium-service zones, representing stable customer behavior and balanced resource allocation; and areas in the lower interval are defined as low-service zones, indicating sparse customer activity and the potential for cost-effective or guided service strategies. This step, through a dynamic grading model based on dwell intensity, transforms customer spatial behavior data into actionable service priorities, providing a quantitative basis for subsequent resource allocation and differentiated operations.

[0141] Develop differentiated operational strategies for high-service, medium-service, and low-service zones;

[0142] Specifically, the process of formulating differentiated operational strategies for high-service, medium-service, and low-service zones involves implementing hierarchical management and strategy matching based on the different priorities of each zone in terms of customer dwell time intensity, resource demand density, and operational objectives. The system first categorizes all openings or seating areas in the target venue into high-service, medium-service, and low-service zones according to the service zone labels generated in the previous steps. For high-service zones, the system sets strategy parameters to prioritize the allocation of marketing resources, human resources, and digital interactive media, while increasing service frequency and interactive response, such as deploying more guides, setting up key advertising placements, and providing real-time guidance information. For medium-service zones, the system sets medium-priority resource investment to ensure basic service quality and makes flexible adjustments based on actual customer flow data, such as periodically placing supplementary advertisements and maintaining adequate guidance manpower coverage. For low-service zones, the system sets basic or economical operational strategies, investing only necessary resources to maintain venue order and provide basic information prompts, while deploying flexible guidance measures to encourage customers to migrate to higher-value areas, thereby optimizing overall operational efficiency and customer distribution structure. The aforementioned differentiation strategy is deployed on the operation platform through preset rules, parameter thresholds, and execution instruction sets, and dynamically adjusted based on real-time data to achieve the operational goal of accurately matching customer needs with service supply.

[0143] Resources are allocated to the high-service, medium-service, and low-service partitions based on the resource quota for each opening.

[0144] Specifically, the process of allocating resources to high-service, medium-service, and low-service zones based on the resource quota for each opening involves a matching mechanism to precisely allocate limited resources to different service levels after prioritizing customer service and calculating the resource quota for each opening. This process first establishes a mapping relationship between the opening and its corresponding service zone, assigning each opening to one of the three service zones (high, medium, or low) based on the opening zone labels formed in previous steps. Then, the resource quota value for each opening is read as the initial allocation for resource scheduling. For openings within a high-service zone, the system sets a resource tilt coefficient with a weighting factor greater than one, proportionally amplifying their corresponding resource quota value to ensure priority access and guarantee that this type of area receives more manpower, digital interactive devices, or marketing placements. For medium-service zones, the resource quota remains unchanged or is adjusted to a stable quota slightly lower than the original value through a balancing coefficient. For low-service zones, the system proportionally reduces the resource quota and centrally allocates it to other higher-priority areas. Ultimately, the system writes the adjusted resource quotas into the partition allocation table, forming a resource distribution plan for regions with different service levels. It then allocates on-site resources through the deployment and execution module to ensure that the operation system improves overall efficiency and service perception quality while meeting customer needs.

[0145] Specifically, high-service zones, medium-service zones, and low-service zones are service demand levels stratified based on customer dwell time data in different areas of the venue. The total dwell time in the stands is a comprehensive indicator reflecting the overall concentration of customer flow, obtained by summing the second cumulative dwell time of all anonymous devices across all open areas. This indicator measures the spatial usage density distribution of the entire venue within a certain timeframe. Service zones are divided based on dwell intensity and resource pressure, prioritizing areas according to customer behavior activity from high to low to support resource allocation and operational strategy formulation. High-service zones represent areas with high customer concentration and intense resource competition, requiring priority for marketing, personnel guidance, and security monitoring. Medium-service zones are areas with medium density, employing a neutral operational strategy. Low-service zones are areas with low customer usage frequency, where energy-saving management or conversion strategies can be adopted to improve utilization. By implementing differentiated operational management for different service zones, refined resource allocation and precise customer management are achieved, enhancing the smart venue operation capabilities driven by big data.

[0146] like Figure 2 The diagram shown is a functional module diagram of a customer management system based on big data provided in an embodiment of the present invention.

[0147] In this embodiment, the functions of each module / unit are as follows:

[0148] The opening association recognition module is used to assign values ​​to the opening association identifier of each seat based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat.

[0149] The structure exposure calculation module is used to determine the structure exposure weight of the Wi-Fi access point to the opening based on the opening association identifier;

[0150] The transmission mapping generation module is used to write the opening transmission table according to the opening association identifier and the structure exposure weight;

[0151] The location duration accounting module is used to calculate the first cumulative dwell time of the anonymous device based on the session log of the target venue and the structure exposure weight in the opening transmission table, and to calculate the second cumulative dwell time.

[0152] The resource share calculation module is used to calculate the resource quota for each opening based on the first cumulative stay time and the total resource amount of the target venue;

[0153] The operations matching and delivery module is used to match operational resources and deliver services to customer segments of the target venue based on the second cumulative stay time and resource quota.

[0154] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A customer management method based on big data, characterized in that, Includes the following steps: S1. Assign values ​​to the opening association identifiers of seats based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat; The specific steps for assigning values ​​to the seat opening association identifier are as follows: Polygon extraction is performed on the opening contour data of the CAD drawing to obtain multiple openings; Extract the edge line from the boundary of the opening; Starting from the direction of the stands, perform a morphological dilation operation on the edge line based on the row spacing of the seats in the target stadium to obtain the shaded area; Determine the spatial overlap between the Wi-Fi access point and the shaded area for each seat; The opening association identifier of the seat is assigned a value based on the judgment result of the spatial overlap relationship. S2. Determine the structural exposure weight of the Wi-Fi access point to the opening based on the opening association identifier; The specific steps for determining the structural exposure weight of a Wi-Fi access point relative to an opening are as follows: Define the surrounding corridor area of ​​the target venue; Discrete ray emission is performed at each Wi-Fi access point; After the discrete ray emission is completed, the number of rays that pass through the opening and reach the corridor area is counted according to the opening association marker; The number of rays is normalized to obtain the structural exposure weight of the Wi-Fi access point to the opening; S3. Write the opening transfer table according to the opening association identifier and the structure exposure weight; S4. Calculate the first cumulative dwell time of the anonymous device based on the session log and the structure exposure weight in the open transmission table of the target venue, and calculate the second cumulative dwell time; The specific steps for calculating the first cumulative dwell time on the anonymous device are as follows: Obtain the session logs of the target venue, which include the anonymous device ID, Wi-Fi access point number, connection start time, and connection end time. The difference between the connection end time and the connection start time is calculated to obtain the dwell time of the anonymous device at the Wi-Fi access point; The dwell time and the structural exposure weight in the aperture transfer table are multiplied to obtain the allocated time of the anonymity device in each aperture direction; The allocated time is accumulated based on the opening number of the opening to obtain the first cumulative dwell time of the anonymous device; The second cumulative dwell time is used to quantify customer activity in areas not covered by openings. S5. Calculate the resource quota for each opening based on the first cumulative stay time and the total resource quantity of the target venue; S6. Based on the second cumulative stay duration and resource quota, perform operational resource matching and service delivery to customer zones of the target venue.

2. The customer management method based on big data according to claim 1, characterized in that, The opening association identifier of the seat is assigned a value based on the judgment result of the spatial overlap relationship, including: If the determination result is that the location coordinates of the Wi-Fi access point fall within the shaded area, then the seat opening association identifier is set to 1; otherwise, the seat opening association identifier is set to 0.

3. The customer management method based on big data according to claim 1, characterized in that, The opening transfer table is written based on the opening association identifier and the structure exposure weight, including: If the Wi-Fi access point is associated with the opening of the seat as 1, then the structural exposure weight of the Wi-Fi access point to the opening is written into the corresponding cell of the opening transmission table. If the Wi-Fi access point's association identifier with the seat opening is 0, then write 0 to the corresponding cell in the opening transmission table.

4. The customer management method based on big data according to claim 1, characterized in that, The calculation of the second cumulative stay includes: The structural exposure weights in the aperture transfer table are summed to obtain the total aperture transfer weights; Subtract the sum of the open transmission weights from 1 to get the remaining proportion of the grandstand area; The time allotted for the grandstand area is obtained by multiplying the dwell time and the remaining proportion of the grandstand area. The duration allocated to the grandstand area is accumulated based on the anonymous device ID to obtain the second cumulative dwell time of the anonymous device.

5. A customer management method based on big data according to claim 1, characterized in that, Based on the first cumulative stay duration and the total resources of the target venue, calculate the resource quota for each opening, including: The operating intensity of each opening is obtained by summing the first cumulative dwell time of all anonymous devices; The total number of openings in the entire site is obtained by summing the operating intensity of all openings. The ratio of operating intensity to the total number of openings in the entire site is calculated to obtain the relative share of each opening; Obtain the total resources of the target venue; The resource allocation value is obtained by multiplying the total resource amount and the relative share of each opening. Round the resource allocation value down to get the resource quota for each opening.

6. A customer management method based on big data according to claim 1, characterized in that, Based on the second cumulative stay duration and resource quota, operational resources are matched and services are deployed to target venues by customer segmentation, including: Sum the second cumulative dwell time of all anonymous devices to obtain the total dwell time in the stands; Based on the total time spent in the stands, customer service needs are prioritized and divided into high-service zones, medium-service zones, and low-service zones. Develop differentiated operational strategies for high-service, medium-service, and low-service zones; Resources are allocated to the high-service, medium-service, and low-service partitions based on the resource quota for each opening.

7. A customer management system based on big data, used to implement the customer management method based on big data as described in any one of claims 1-6, characterized in that, The system includes: The opening association recognition module is used to assign values ​​to the opening association identifier of each seat based on the opening outline data of the CAD drawing in the target venue and the Wi-Fi access point of each seat. The structure exposure calculation module is used to determine the structure exposure weight of the Wi-Fi access point to the opening based on the opening association identifier; The transmission mapping generation module is used to write the opening transmission table according to the opening association identifier and the structure exposure weight; The location duration accounting module is used to calculate the first cumulative dwell time of the anonymous device based on the session log of the target venue and the structure exposure weight in the opening transmission table, and to calculate the second cumulative dwell time. The resource share calculation module is used to calculate the resource quota for each opening based on the first cumulative stay time and the total resource amount of the target venue; The operations matching and delivery module is used to match operational resources and deliver services to customer segments of the target venue based on the second cumulative stay time and resource quota.

Images