An online tracking, training, prediction and analysis system for customer flow in large shopping malls and stores

Abstract

The present invention provides a system for online tracking, training, prediction and analysis of passenger flow in large shopping malls and stores, which belongs to the field of passenger flow analysis in shopping malls. It solves the problems that the existing passenger flow analysis system has a single function and cannot realize online tracking, training, prediction and personalized services of passenger flow. It includes a data acquisition layer, a data transmission layer, a data analysis layer, a data storage layer, a real-time passenger flow monitoring and tracking subsystem, a behavior analysis subsystem, a comparative analysis subsystem, a customer group segmentation subsystem, a personal profile subsystem, a trend prediction subsystem, a data visualization subsystem, a store marketing decision subsystem, a rental decision subsystem, an anomaly detection subsystem, a simulation training subsystem and a shopping mall software subsystem. The data transmission layer is electrically connected to the shopping mall cloud platform and terminal equipment. The present invention can efficiently perform data acquisition, transmission, analysis and storage, realize online tracking, training, prediction and simulation optimization of passenger flow, plan navigation routes and select products, and provide personalized services.

Description

Technical Field

[0001] The invention belongs to the technical field of passenger flow analysis in large shopping malls, and in particular to an online tracking, training, prediction and analysis system for passenger flow in stores in large shopping malls. Background Art

[0002] With the rapid development of social economy and the continuous improvement of people's living standards, large shopping malls have become one of the main places for people to entertain themselves. During holidays, the passenger flow in large shopping malls will be more concentrated. Precision marketing to precise customers based on historical data will attract more customers to the mall for activities.

[0003] Currently, the analysis of shopping mall customer flow is mostly based on the amount of counter consumption combined with simple differentiation analysis during holidays, but the analysis of information such as customer source distribution, surrounding environment, and dwelling time is relatively limited. Therefore, shopping malls do not have objective data based on this for traffic diversion.

[0004] In particular, if the specific situation of the commodity order is unclear, a large number of customers arriving at the store can easily cause congestion in the store and passers-by, and if the order is not clear in the mall, it is easy for customers to miss the big order;

[0005] How to track customer flow online and conduct training predictions, perform simulation optimization of shopping mall services, design new categories of products or services after multiple analyses of self-training and learning, plan navigation routes and select products, and provide personalized services are issues that need to be addressed urgently.

[0006] Therefore, we propose an online tracking, training, prediction and analysis system for store traffic in large shopping malls, which can efficiently collect, transmit, analyze and store data, realize online tracking, training, prediction and simulation optimization of customer flow, plan navigation routes and select products, and provide personalized services. Summary of the invention

[0007] The purpose of the present invention is to address the above-mentioned problems in the existing technology and to propose an online tracking, training, prediction and analysis system for the store traffic in large shopping malls. The technical problem to be solved by the invention is: how to achieve efficient data collection, transmission, analysis and storage, realize online tracking, training, prediction and simulation optimization of customer traffic, plan navigation routes and select products, and provide personalized services.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] A system for online tracking, training, prediction and analysis of store customer flow in a large shopping mall comprises a data collection layer, a data transmission layer, a data analysis layer, a data storage layer, a real-time customer flow monitoring and tracking subsystem, a behavior analysis subsystem, a comparative analysis subsystem, a customer group segmentation subsystem, a personal profile subsystem, a trend prediction subsystem, a data visualization subsystem, a store marketing decision subsystem, a rental decision subsystem, an anomaly detection subsystem, a simulation training subsystem and a shopping mall software subsystem. The data transmission layer is electrically connected to a shopping mall cloud platform and a terminal device, and the shopping mall software subsystem is arranged on the terminal device.

[0010] The working principle of the present invention is: the data acquisition layer is used to collect data from various parking lots, entrances and exits, floor passages, elevator entrances and store entrances in real time, and the collected data is accurately transmitted to the data processing center or server in real time through the data transmission layer in the form of a wired or wireless network. The data analysis layer uses data analysis algorithms and models to process and analyze the collected data, and the data storage layer uses big data storage technology to store massive passenger flow data, providing powerful computing power for data storage and processing.

[0011] Real-time customer flow monitoring and tracking, making real-time heat maps, intuitively displaying customer flow density and flow trends, analyzing customer movement paths and dwelling times, identifying customer points of interest and shopping preferences, conducting comparative analysis of customer flow in different stores, different floors and different time periods, evaluating the operating efficiency of each store or region, identifying potential problems and opportunities, and formulating differentiated market expansion strategies;

[0012] It can segment customer groups, establish personal profiles, make trend forecasts, make store marketing decisions, and detect anomalies to prepare response measures in advance;

[0013] Through several simulation robots, shopping mall services are simulated and optimized, and after multiple analyses of self-training and learning, new categories of products or services are designed;

[0014] The data visualization subsystem displays data and analysis results in real time, allowing managers to intuitively understand passenger flow conditions;

[0015] Users can plan the best 3D navigation route, select products, and push advertisements through the mall software subsystem on their terminal devices according to their needs. They can also collect and summarize all information using the mall software subsystem and transmit it to the data analysis layer for big data analysis to form the most accurate personalized service.

[0016] The data collection layer is to install advanced sensor equipment and Wi-Fi networks as well as membership card or mobile application data in key locations of the shopping mall, including but not limited to parking lots, entrances and exits, floor passages, elevator entrances and store entrances; the sensor equipment includes but is not limited to face recognition USB cameras, AI cameras, infrared sensors, laser sensors, video surveillance cameras, 3D ToF passenger flow meters, binocular stereo vision sensors and passenger flow statistics cameras.

[0017] With the above structure, the sensor equipment collects data more accurately and with a high degree of automation, can monitor continuously for a long time, and can be integrated with other systems;

[0018] By using the Wi-Fi network in the venue, the Wi-Fi signal of the user's mobile device is detected to count the passenger flow and user stay time. After the user enters the venue, his mobile device will automatically search and connect to the Wi-Fi network, and the Wi-Fi probe device can capture the device information and perform data analysis; while collecting passenger flow data, it can also obtain some user device related information, such as device type, MAC address, etc.

[0019] For merchants with membership systems or mobile applications, through member registration information, login records, consumption records and in-app positioning functions, they can accurately analyze member customers' traffic flow, consumption behavior, store visit frequency and other information, and achieve personalized marketing and service push.

[0020] The data transmission layer transmits the collected data to the data processing center or server in real time through a wired or wireless network to ensure timely and accurate transmission of the data. Wireless network communication technologies include but are not limited to Wi-Fi, Bluetooth and Zigbee, and wired network communication technologies include but are not limited to Ethernet. The data is sent to the back-end server for storage and analysis.

[0021] With the above structure, the wired network has the advantages of fast transmission speed and high stability; the wireless network also makes it convenient for customers to provide feedback information through the mall's APP or mini-programs, enriching the data source; the two methods work stably to ensure timely and accurate transmission of data.

[0022] The data analysis layer uses data analysis algorithms and models to process and analyze the collected data and extract valuable information, including but not limited to the peak and trough periods of customer flow, the age and gender distribution of customers, and customer preferences; data analysis methods include but are not limited to the enterprise-level big data computing platform built by Apache Flink, data mining, machine learning, statistical analysis and AI analysis systems.

[0023] The above structure is used for real-time data analysis, and the business district is analyzed from multiple data sources and multiple dimensions.

[0024] The data storage layer uses big data storage technology to store massive amounts of passenger flow data. The storage technology includes but is not limited to distributed file systems and distributed databases. The data storage has high reliability and scalability to cope with the ever-increasing amount of data, and provides powerful computing power for data storage and processing, ensuring that the system can quickly respond to user queries and analysis requests, and achieve real-time updates and dynamic display of data.

[0025] The real-time passenger flow monitoring and tracking subsystem extracts and identifies human features, combines multiple features, and effectively solves the problem of difficulty in distinguishing boundaries due to the overlap of different people in crowded scenes. It can easily identify multiple human bodies and count them. It can monitor the changes in passenger flow in the mall and each store in real time. The changes in passenger flow include but are not limited to the number of people entering and leaving, floor distribution, store visits, and the concentration of hot spots, and create real-time heat maps to intuitively display the density and flow trends of passenger flow.

[0026] The above structure can monitor the changes in customer flow in shopping malls and stores in real time, allowing managers to grasp the mall's operational dynamics at any time and adjust resource allocation and marketing strategies in a timely manner.

[0027] The behavior analysis subsystem analyzes customers' movement paths and dwell time, identifies customers' points of interest and shopping preferences, helps merchants optimize store layout and product display, and combines POS system data to conduct shopping basket analysis to understand customers' purchasing habits and preferences; the behavior analysis content includes but is not limited to dwell time, walking paths, points of interest, browsing status and purchasing status.

[0028] The comparative analysis subsystem conducts comparative analysis on the customer flow of different stores, different floors and different time periods, evaluates the operating efficiency of each store or area, discovers potential problems and opportunities, provides reference for the overall planning and layout adjustment of the shopping mall, and optimizes the store layout; finds out the differences and reasons for changes, and provides a basis for decision-making; compares the customer flow of chain stores in different cities or regions, understands the consumption potential and competitive situation of different regional markets, and formulates differentiated market expansion strategies; the comparative analysis indicators include but are not limited to customer flow, average stay time, peak and trough time periods, ratio of new and old customers, customer activity, customer visit frequency, repeat customer identification and member identification.

[0029] The customer group segmentation subsystem uses AI technology to identify customers' age, gender, shopping history, spending power and behavioral characteristics, build customer portraits, understand the characteristics and needs of different customer groups, and segment customer groups into groups including but not limited to high-value customers and potential customers, providing a basis for personalized marketing and precise push notifications, and improving customer satisfaction and loyalty.

[0030] The personal profile subsystem uses facial recognition technology to detect the user's facial portrait and establishes a personal profile based on facial attributes, which include gender, age and expression; personal profile information includes but is not limited to gender, age, entry and exit time, walking plan route map, 3D dynamic route map, stay time in each store, purchased product categories, consumption amount, payment method and customer source.

[0031] The above structure helps managers understand consumers' consumption habits and needs, provides data support for subsequent marketing strategy formulation, and ensures rapid tracking and analysis when the user enters the mall again or uses the mall software subsystem again.

[0032] The trend prediction subsystem predicts customer flow in a specific period based on historical data and machine learning algorithms to help shopping malls prepare in advance. The specific period includes but is not limited to holidays, weekends, seasons, economic environment and social events; preparation matters include but are not limited to adjusting staff allocation, arranging staff shifts, adjusting commodity inventory, adjusting commodity sales strategies, increasing investment in tourism facilities and strengthening safety management; commodity sales strategies include but are not limited to pushing coupons, precise delivery and recommending personalized advertisements.

[0033] The data visualization subsystem presents the customer flow data in the form of chart reports, making the data more intuitive, easy to understand and analyze, and displaying it in real time on an electronic billboard or LCD screen, so that managers can intuitively understand the customer flow situation; chart reports include but are not limited to daily / weekly / monthly customer flow reports, customer behavior analysis reports, customer flow forecast analysis reports, holiday customer flow forecast reports, seasonal customer flow forecast reports, customer portrait and segmentation reports, store operation optimization reports, staffing optimization reports and comprehensive evaluation and improvement suggestion reports.

[0034] The store marketing decision subsystem provides accurate customer flow data and analysis reports to support the formulation and implementation of marketing decisions. Based on customer flow data and customer behavior analysis, it evaluates the store's attractiveness and operating efficiency, conducts comparative analysis with similar stores, identifies gaps and areas for improvement, and makes store optimization suggestions. The optimization suggestions include but are not limited to adjusting shelf positions, optimizing store merchandise display and layout, and optimizing traffic flow design, so as to improve customers' shopping experience and purchase conversion rate.

[0035] The leasing decision subsystem provides data support for the investment promotion and leasing departments of the shopping mall. It reasonably determines the rental level and lease term of the store based on the passenger flow and customer characteristics information in different areas; evaluates the potential value and risk of new stores to provide a reference for investment promotion decisions; and provides a reference for commercial site selection. For companies that are opening new stores or expanding their businesses, it analyzes the passenger flow and its characteristics around potential sites, and can predict the commercial potential of the location and the business formats suitable for operation.

[0036] The anomaly detection subsystem identifies those high or low traffic points that deviate from the normal ones. For the detected anomalies, the system will conduct an in-depth analysis of the relevant factors. Based on the anomaly analysis results, the system can adjust the prediction model. If continued anomalies are expected, the system may issue an alarm to the mall managers so that they can prepare countermeasures in advance; abnormal events include but are not limited to temporary events, seasonal changes and system failures.

[0037] The simulation training subsystem sets up several simulation robots to simulate real gender, age, occupation and family status, and conducts simulation optimization of shopping mall services. It simulates all shopping mall service needs of the simulation robots, sets up new categories of products or services in shopping malls and stores, conducts simulation tests, analyzes self-training and learning many times, designs new categories of products or services, and then analyzes and optimizes them according to actual conditions. The results are entered into the personal files of the simulation robots, and summary statistics and predictive analysis are performed. Finally, simulation training is performed on the predictive analysis results and data mining projects.

[0038] The mall software subsystem is but not limited to a mall APP, a mall applet or a mall official account; the functions of the mall software subsystem include but are not limited to a 3D positioning navigation map, a navigation route planning module, a related product recommendation module, an interest product recommendation module, an advertising push module, an intelligent route recommendation module and a data collection module; the locations of the 3D positioning navigation map include but are not limited to parking lot locations, parking space locations, shop locations, product locations, restroom locations, smoking areas and rest areas; the navigation route planning plans the optimal 3D navigation route between the user's entry location into the mall and the designated arrival location, or plans the optimal 3D navigation route based on a number of products selected by the user or the places to be visited, and can be changed at any time according to the user's needs, and the route can be reconfirmed. The optimal 3D navigation route includes the shortest distance mode, indoor parking lot mode, outdoor parking lot mode, lowest price mode and system recommendation mode;

[0039] The related product recommendation module recommends products on the planned 3D navigation route to facilitate users to purchase;

[0040] The product recommendation module recommends products of interest based on the user's purchase history data, improving the customer's shopping experience and purchase conversion rate;

[0041] The advertising push module pushes personalized promotional information and coupons based on the customer's shopping history and behavior data;

[0042] The intelligent route recommendation module intelligently recommends several routes of interest based on the user's shopping history and navigation route history, including but not limited to food routes, shopping routes or overlapping routes;

[0043] The data collection module will use all the information collected by the mall software subsystem and transmit it to the data analysis layer for big data analysis to obtain more accurate data, enter it into personal files, and form the most accurate personalized service.

[0044] With the above structure, before going to a large shopping mall, the user uses the mall software subsystem to select the parking lot location, parking space location, store location, and commodity location through the 3D positioning navigation map; the navigation route is planned to plan the optimal 3D navigation route between the user's entry into the mall and the designated destination, or the optimal 3D navigation route is planned according to the user's selected commodities or places to go. The user can choose the optimal 3D navigation route according to their needs: shortest distance mode, indoor parking mode, outdoor parking mode, lowest price mode, and system recommended mode; the navigation route is planned to plan the optimal 3D navigation route between the user's entry into the mall and the designated destination, or the optimal 3D navigation route is planned according to the user's selected commodities or places to go. The route can be changed and reconfirmed at any time according to the needs of users. The optimal 3D navigation route includes the shortest distance mode, indoor parking mode, outdoor parking mode, lowest price mode and system recommendation mode; users can purchase the products on the planned 3D navigation route recommended by the related product recommendation module; users can purchase the products of interest recommended by the interest product recommendation module; users can make choices based on the personalized promotional information and coupons pushed by the advertising push module; users can make route choices based on the intelligent route recommendation module that intelligently recommends several routes of interest: dining routes, shopping routes or overlapping routes; after the user makes a selection, all data information will be collected and summarized by the data collection module and transmitted to the data analysis layer for big data analysis to obtain more accurate data.

[0045] Compared with the existing technology, this large-scale shopping mall store customer flow online tracking training prediction analysis system has the following advantages:

[0046] The data collection layer is used to collect data from various parking lots, entrances and exits, floor passages, elevator entrances and store entrances in real time. The collected data is accurately transmitted to the data processing center or server in real time through the data transmission layer in the form of wired or wireless networks. The data analysis layer uses data analysis algorithms and models to process and analyze the collected data. The data storage layer uses big data storage technology to store massive passenger flow data, providing powerful computing power for data storage and processing.

[0047] Real-time customer flow monitoring and tracking, making real-time heat maps, intuitively displaying customer flow density and flow trends, analyzing customer movement paths and dwelling times, identifying customer points of interest and shopping preferences, conducting comparative analysis of customer flow in different stores, different floors and different time periods, evaluating the operating efficiency of each store or region, identifying potential problems and opportunities, and formulating differentiated market expansion strategies;

[0048] It can segment customer groups, establish personal profiles, make trend forecasts, make store marketing decisions, and detect anomalies to prepare response measures in advance;

[0049] Through several simulation robots, shopping mall services are simulated and optimized, and after multiple analyses of self-training and learning, new categories of products or services are designed;

[0050] The data visualization subsystem displays data and analysis results in real time, allowing managers to intuitively understand passenger flow conditions;

[0051] Users can plan the best 3D navigation route, select products, and push advertisements through the mall software subsystem according to their needs. All information collected and summarized by the mall software subsystem will be transmitted to the data analysis layer for big data analysis to form the most accurate personalized service. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is the overall system block diagram of the present invention.

[0053] Figure 2 It is a system block diagram of the data acquisition layer and the data transmission layer in the present invention.

[0054] Figure 3 It is a system block diagram of the real-time passenger flow monitoring and tracking subsystem in the present invention.

[0055] Figure 4 It is a system block diagram of the behavior analysis subsystem in the present invention.

[0056] Figure 5 It is a system block diagram of the comparative analysis sub-system spinning in the present invention.

[0057] Figure 6 It is a system block diagram of the customer group segmentation subsystem in the present invention.

[0058] Figure 7 It is a system block diagram of the personal profile subsystem in the present invention.

[0059] Figure 8 It is a system block diagram of the trend prediction subsystem in the present invention.

[0060] Fig. 9 It is a system block diagram of the data visualization subsystem in the present invention.

[0061] Fig.10 It is a system block diagram of the anomaly detection subsystem in the present invention.

[0062] Fig.11 It is a system block diagram of the shopping mall software subsystem in the present invention. DETAILED DESCRIPTION

[0063] The following are specific embodiments of the present invention and the accompanying drawings to further describe the technical solution of the present invention, but the present invention is not limited to these embodiments.

[0064] like Figure 1-Figure 11 As shown, the online tracking, training, prediction and analysis system for store customer flow in a large shopping mall includes a shopping mall cloud platform, a data collection layer, a data transmission layer, a data analysis layer, a data storage layer, a real-time customer flow monitoring and tracking subsystem, a behavior analysis subsystem, a comparative analysis subsystem, a customer group segmentation subsystem, a personal profile subsystem, a trend prediction subsystem, a data visualization subsystem, a store marketing decision subsystem, a leasing decision subsystem, an anomaly detection subsystem, a simulation training subsystem and a shopping mall software subsystem. The shopping mall cloud platform is electrically connected to the data transmission layer and the terminal device respectively, and the shopping mall software subsystem is arranged on the terminal device.

[0065] The data collection layer is used to collect data from various parking lots, entrances and exits, floor passages, elevator entrances and store entrances in real time. The collected data is accurately transmitted to the data processing center or server in real time through the data transmission layer in the form of wired or wireless networks. The data analysis layer uses data analysis algorithms and models to process and analyze the collected data. The data storage layer uses big data storage technology to store massive passenger flow data, providing powerful computing power for data storage and processing.

[0066] Real-time customer flow monitoring and tracking, making real-time heat maps, intuitively displaying customer flow density and flow trends, analyzing customer movement paths and dwelling times, identifying customer points of interest and shopping preferences, conducting comparative analysis of customer flow in different stores, different floors and different time periods, evaluating the operating efficiency of each store or region, identifying potential problems and opportunities, and formulating differentiated market expansion strategies;

[0067] It can segment customer groups, establish personal profiles, make trend forecasts, make store marketing decisions, and detect anomalies to prepare response measures in advance;

[0068] Through several simulation robots, shopping mall services are simulated and optimized, and after multiple analyses of self-training and learning, new categories of products or services are designed;

[0069] The data visualization subsystem displays data and analysis results in real time, allowing managers to intuitively understand passenger flow conditions;

[0070] Users can plan the best 3D navigation route, select products, and push advertisements through the mall software subsystem on their terminal devices according to their needs. They can also collect and summarize all information using the mall software subsystem and transmit it to the data analysis layer for big data analysis to form the most accurate personalized service.

[0071] The data collection layer is to install advanced sensor equipment and Wi-Fi networks as well as membership card or mobile application data in key locations of the shopping mall, but not limited to parking lots, entrances and exits, floor passages, elevator entrances and store doors; sensor equipment includes but is not limited to face recognition USB cameras, AI cameras, infrared sensors, laser sensors, video surveillance cameras, 3DToF passenger flow meters, binocular stereo vision sensors and passenger flow statistics cameras.

[0072] Sensor equipment collects data more accurately and with a high degree of automation, can monitor continuously for a long time, and can be integrated with other systems;

[0073] By using the Wi-Fi network in the venue, the Wi-Fi signal of the user's mobile device is detected to count the passenger flow and user stay time. After the user enters the venue, his mobile device will automatically search and connect to the Wi-Fi network, and the Wi-Fi probe device can capture the device information and perform data analysis; while collecting passenger flow data, it can also obtain some user device related information, such as device type, MAC address, etc.

[0074] For merchants with membership systems or mobile applications, through member registration information, login records, consumption records and in-app positioning functions, they can accurately analyze member customers' traffic flow, consumption behavior, store visit frequency and other information, and achieve personalized marketing and service push.

[0075] The data transmission layer transmits the collected data to the data processing center or server in real time through a wired or wireless network to ensure timely and accurate data transmission. Wireless network communication technologies include but are not limited to Wi-Fi, Bluetooth and Zigbee, and wired network communication technologies include but are not limited to Ethernet, which sends the data to the back-end server for storage and analysis. Wired networks have the advantages of fast transmission speed and high stability; wireless networks also facilitate customers to provide feedback information through the mall's APP or mini-programs, enriching data sources; the two methods work stably together to ensure timely and accurate data transmission.

[0076] The data analysis layer uses data analysis algorithms and models to process and analyze the collected data and extract valuable information, including but not limited to the peak and trough periods of customer traffic, the age and gender distribution of customers, and customer preferences; data analysis methods include but are not limited to the enterprise-level big data computing platform built by Apache Flink, data mining, machine learning, statistical analysis, and AI analysis systems. Real-time data analysis, multi-data source, multi-dimensional analysis of the business district.

[0077] The data storage layer uses big data storage technology to store massive amounts of passenger flow data. Storage technology includes but is not limited to distributed file systems and distributed databases. Data storage has high reliability and scalability to cope with the ever-increasing amount of data. It provides powerful computing power for data storage and processing, ensuring that the system can quickly respond to user queries and analysis requests, and achieve real-time updates and dynamic display of data.

[0078] The real-time passenger flow monitoring and tracking subsystem extracts and identifies human features, and combines multiple features to effectively solve the problem of difficult-to-distinguish boundaries caused by overlapping people in crowded scenes. It can easily identify multiple human bodies and count them. It can monitor the changes in passenger flow in the mall and each store in real time. Passenger flow changes include but are not limited to the number of people entering and leaving, floor distribution, store visits, and hot spot area aggregation. It can create real-time heat maps to intuitively display the density and flow trends of passenger flow. It can monitor the changes in passenger flow in the mall and stores in real time, allowing managers to keep abreast of the mall's operating dynamics at any time and adjust resource allocation and marketing strategies in a timely manner.

[0079] The behavior analysis subsystem helps merchants optimize store layout and product display by analyzing customers' movement paths and dwell time, identifying customers' points of interest and shopping preferences. It also conducts shopping basket analysis in combination with POS system data to understand customers' purchasing habits and preferences. The behavior analysis content includes but is not limited to dwell time, walking paths, points of interest, browsing behavior and purchasing behavior.

[0080] The comparative analysis subsystem conducts comparative analysis on the customer flow of different stores, different floors and different time periods, evaluates the operating efficiency of each store or area, discovers potential problems and opportunities, provides reference for the overall planning and layout adjustment of the shopping mall, and optimizes the store layout; finds out the differences and reasons for changes, and provides a basis for decision-making; compares the customer flow of chain stores in different cities or regions, understands the consumption potential and competitive situation of different regional markets, and formulates differentiated market expansion strategies; comparative analysis indicators include but are not limited to customer flow, average stay time, peak and trough time periods, ratio of new and old customers, customer activity, customer visit frequency, repeat customer identification and member identification.

[0081] The customer segmentation subsystem uses AI technology to identify customers' characteristics, including but not limited to age, gender, shopping history, spending power and behavioral characteristics, build customer portraits, understand the characteristics and needs of different customer groups, and segment customer groups into groups including but not limited to high-value customers and potential customers, providing a basis for personalized marketing and precise push notifications, and improving customer satisfaction and loyalty.

[0082] The personal profile subsystem uses face recognition technology to detect the user's face portrait and establishes a personal profile based on face attributes, including gender, age and expression; personal profile information includes but is not limited to gender, age, entry and exit time, walking plan route map, 3D dynamic route map, time spent in each store, purchased product categories, consumption amount, payment method and customer source. It helps managers understand consumers' consumption habits and needs, provides data support for subsequent marketing strategy formulation, and ensures rapid tracking and analysis when the user enters the mall again or uses the mall software subsystem again.

[0083] The trend prediction subsystem predicts customer flow in a specific period based on historical data and machine learning algorithms to help shopping malls prepare in advance. Specific periods include but are not limited to holidays, weekends, seasons, economic environments and social events; preparations include but are not limited to adjusting staffing, arranging staff shifts, adjusting product inventory, adjusting product sales strategies, increasing investment in tourism facilities and strengthening safety management; product sales strategies include but are not limited to pushing coupons, precise delivery and recommending personalized advertisements.

[0084] The data visualization subsystem presents the customer flow data in the form of chart reports, making the data more intuitive, easy to understand and analyze, and displaying it in real time on electronic billboards or LCD screens, so that managers can intuitively understand the customer flow situation; chart reports include but are not limited to daily / weekly / monthly customer flow reports, customer behavior analysis reports, customer flow forecast analysis reports, holiday customer flow forecast reports, seasonal customer flow forecast reports, customer portrait and segmentation reports, store operation optimization reports, staffing optimization reports and comprehensive evaluation and improvement suggestion reports.

[0085] The store marketing decision subsystem provides accurate customer flow data and analysis reports to support the formulation and implementation of marketing decisions. Based on customer flow data and customer behavior analysis, it evaluates the store's attractiveness and operating efficiency, conducts comparative analysis with similar stores, identifies gaps and areas for improvement, and makes store optimization suggestions. The optimization suggestions include but are not limited to adjusting shelf positions, optimizing store merchandise display and layout, and optimizing traffic flow design, thereby improving customers' shopping experience and purchase conversion rates.

[0086] The leasing decision subsystem provides data support for the mall's investment and leasing departments. It reasonably determines the rent level and lease term of stores based on the passenger flow and customer characteristics of different areas. It evaluates the potential value and risks of new stores to provide a reference for investment decisions. It provides a reference for commercial site selection. For companies that are opening new stores or expanding their businesses, it analyzes the passenger flow and its characteristics around potential sites, and can predict the commercial potential of the location and the business formats that are suitable for operation.

[0087] The anomaly detection subsystem identifies those high or low traffic points that deviate from the normal ones. For the detected anomalies, the system will conduct an in-depth analysis of the relevant factors. Based on the anomaly analysis results, the system can adjust the prediction model. If continued anomalies are expected, the system may issue an alarm to the mall managers so that they can prepare countermeasures in advance; abnormal events include but are not limited to temporary events, seasonal changes and system failures.

[0088] The simulation training subsystem sets up several simulation robots to simulate real gender, age, occupation and family status, and conducts simulation optimization of shopping mall services. It simulates all shopping mall service needs of the simulation robots, sets up new categories of products or services in shopping malls and stores, conducts simulation tests, and designs new categories of products or services after multiple analyses of self-training and learning. It then analyzes and optimizes based on actual conditions, enters the results into the personal files of the simulation robots, conducts summary statistics and predictive analysis, and finally conducts simulation training on the predictive analysis results and data mining projects.

[0089] The mall software subsystem includes but is not limited to the mall APP, mall mini program or mall public account; the functions of the mall software subsystem include but are not limited to 3D positioning navigation map, planning navigation route module, related product recommendation module, interest product recommendation module, advertising push module and intelligent route recommendation module and data collection module; the location of the 3D positioning navigation map is not limited to the parking lot location, parking space location, store location, product location, toilet location, smoking area and rest area; the planning navigation route plans the optimal 3D navigation route between the user's entry location and the designated arrival location or plans the optimal 3D navigation route based on the user's selected products or places to go, and can be changed at any time according to the user's needs and the route can be reconfirmed. The optimal 3D navigation route includes the shortest distance mode, indoor parking lot mode, outdoor parking lot mode, lowest price mode and system recommendation mode;

[0090] The related product recommendation module recommends products on the planned 3D navigation route to facilitate users to purchase;

[0091] The product recommendation module recommends products of interest based on the user's purchase history data, improving the customer's shopping experience and purchase conversion rate;

[0092] The advertising push module pushes personalized promotional information and coupons based on the customer's shopping history and behavior data;

[0093] The intelligent route recommendation module intelligently recommends several routes of interest based on the user's shopping history and navigation route history, including but not limited to food routes, shopping routes or overlapping routes;

[0094] The data collection module will use all the information collected by the mall software subsystem and transmit it to the data analysis layer for big data analysis to obtain more accurate data, enter it into personal files, and form the most accurate personalized service.

[0095] Before going to a large shopping mall, the user uses the mall software subsystem to select the parking lot location, parking space location, store location, and product location through the 3D positioning navigation map; the navigation route is planned to plan the optimal 3D navigation route between the user's entry into the mall and the designated destination, or the optimal 3D navigation route is planned based on the user's selected products or places to go. Users can choose the optimal 3D navigation route according to their needs: shortest distance mode, indoor parking mode, outdoor parking mode, lowest price mode, and system recommended mode; the navigation route is planned to plan the optimal 3D navigation route between the user's entry into the mall and the designated destination, or the optimal 3D navigation route is planned based on the user's selected products or places to go, and the user can choose the optimal 3D navigation route according to their needs: shortest distance mode, indoor parking mode, outdoor parking mode, lowest price mode, and system recommended mode. The user's needs change at any time and the route needs to be reconfirmed. The optimal 3D navigation route includes the shortest distance mode, indoor parking mode, outdoor parking mode, lowest price mode and system recommendation mode; the user selects and purchases the products on the planned 3D navigation route based on the recommendations of the associated product recommendation module; the user selects and purchases the products of interest based on the recommendations of the interest product recommendation module; the user selects the personalized promotional information and coupons pushed by the advertising push module; the user selects the route based on the intelligent route recommendation module that intelligently recommends several routes of interest: dining routes, shopping routes or overlapping routes; after the user makes a selection, all data information is collected and summarized by the data collection module and transmitted to the data analysis layer for big data analysis to obtain more accurate data.

[0096] In summary, the data collection layer is used to collect data from each parking lot, entrance and exit, floor passage, elevator entrance and store entrance in real time, and the collected data is accurately transmitted to the data processing center or server in real time through the data transmission layer in the form of wired or wireless network. The data analysis layer uses data analysis algorithms and models to process and analyze the collected data, and the data storage layer uses big data storage technology to store massive passenger flow data, providing powerful computing power for data storage and processing.

[0097] Real-time customer flow monitoring and tracking, making real-time heat maps, intuitively displaying customer flow density and flow trends, analyzing customer movement paths and dwelling times, identifying customer points of interest and shopping preferences, conducting comparative analysis of customer flow in different stores, different floors and different time periods, evaluating the operating efficiency of each store or region, identifying potential problems and opportunities, and formulating differentiated market expansion strategies;

[0098] It can segment customer groups, establish personal profiles, make trend forecasts, make store marketing decisions, and detect anomalies to prepare response measures in advance;

[0099] Through several simulation robots, shopping mall services are simulated and optimized, and after multiple analyses of self-training and learning, new categories of products or services are designed;

[0100] The data visualization subsystem displays data and analysis results in real time, allowing managers to intuitively understand passenger flow conditions;

[0101] Users can plan the best 3D navigation route, select products, and push advertisements through the mall software subsystem according to their needs. All information collected and summarized by the mall software subsystem will be transmitted to the data analysis layer for big data analysis to form the most accurate personalized service.

[0102] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. A large-scale shopping mall store customer flow online tracking training prediction analysis system, characterized in that: It includes data collection layer, data transmission layer, data analysis layer, data storage layer, real-time customer flow monitoring and tracking subsystem, behavior analysis subsystem, comparative analysis subsystem, customer group segmentation subsystem, personal profile subsystem, trend prediction subsystem, data visualization subsystem, store marketing decision subsystem, leasing decision subsystem, anomaly detection subsystem, simulation training subsystem and shopping mall software subsystem. The data transmission layer is electrically connected to the shopping mall cloud platform and terminal equipment, and the shopping mall software subsystem is arranged on the terminal equipment.

2. According to claim 1, a large-scale shopping mall store customer flow online tracking training prediction analysis system is characterized in that: The data collection layer is to install advanced sensor equipment and Wi-Fi networks as well as membership card or mobile application data in key locations of the shopping mall, but not limited to parking lots, entrances and exits, floor passages, elevator entrances and store entrances; the sensor equipment is but not limited to face recognition USB cameras, AI cameras, infrared sensors, laser sensors, video surveillance cameras, 3D ToF passenger flow meters, binocular stereo vision sensors and passenger flow statistics cameras; the data transmission layer transmits the collected data to the data processing center or server in real time through a wired or wireless network to ensure timely and accurate transmission of the data. Wireless network communication technologies include but are not limited to Wi-Fi, Bluetooth and Zigbee, and wired network communication technologies include but are not limited to Ethernet, and the data is sent to the back-end server for storage and analysis.

3. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 2, characterized in that: The data analysis layer uses data analysis algorithms and models to process and analyze the collected data and extract valuable information, including but not limited to the peak and trough periods of customer flow, the age and gender distribution of customers, and customer preferences; the data analysis methods include but are not limited to the enterprise-level big data computing platform built by Apache Flink, data mining, machine learning, statistical analysis and AI analysis systems; the data storage layer uses big data storage technology to store massive amounts of customer flow data, and the storage technology includes but is not limited to distributed file systems and distributed databases. The data storage has high reliability and scalability to cope with the growing amount of data, and provides powerful computing power for data storage and processing, ensuring that the system can quickly respond to user queries and analysis requests, and realize real-time update and dynamic display of data.

4. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 3, characterized in that: The real-time passenger flow monitoring and tracking subsystem extracts and identifies human features, combines multiple features, and effectively solves the problem of difficulty in distinguishing boundaries due to the overlap of different people in crowded scenes. It can easily identify multiple human bodies and count them; it can monitor the changes in passenger flow in the mall and each store in real time. The changes in passenger flow include but are not limited to the number of people entering and leaving, floor distribution, store visits, and the concentration of hot spots, and create real-time heat maps to intuitively display the density and flow trend of passenger flow; the behavior analysis subsystem analyzes the customer's movement path and residence time, identifies the customer's points of interest and shopping preferences, helps merchants optimize store layout and product display, and combines POS system data to perform shopping basket analysis to understand customers' purchasing habits and preferences; the content of behavior analysis is but not limited to residence time, walking path, points of interest, browsing situation and purchase situation.

5. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 4, characterized in that: The comparative analysis subsystem conducts comparative analysis on the customer flow of different stores, different floors and different time periods, evaluates the operating efficiency of each store or area, discovers potential problems and opportunities, provides reference for the overall planning and layout adjustment of the shopping mall, and optimizes the store layout; finds out the differences and reasons for changes, and provides a basis for decision-making; compares the customer flow of chain stores in different cities or regions, understands the consumption potential and competitive situation of different regional markets, and formulates differentiated market expansion strategies; the comparative analysis indicators include but are not limited to customer flow, average stay time, peak and trough time periods, ratio of new and old customers, customer activity, customer visit frequency, repeat customer identification and member identification.

6. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 5, characterized in that: The customer group segmentation subsystem uses AI technology to identify customers' age, gender, shopping history, spending power and behavioral characteristics, build customer portraits, understand the characteristics and needs of different customer groups, and segment customer groups into groups including but not limited to high-value customers and potential customers, providing a basis for personalized marketing and precise push notifications, and improving customer satisfaction and loyalty; the personal profile subsystem uses face recognition technology to detect the user's face portrait and establish a personal profile based on face attributes, which include gender, age and expression; personal profile information includes but is not limited to gender, age, entry and exit time, walking plan route map, 3D dynamic route map, stay time in each store, purchased product categories, consumption amount, payment method and customer source.

7. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 6, characterized in that: The trend prediction subsystem predicts the customer flow in a specific period based on historical data and machine learning algorithms to help shopping malls prepare in advance. The specific period includes but is not limited to holidays, weekends, seasons, economic environment and social events; preparation matters include but are not limited to adjusting staffing, arranging staff shifts, adjusting commodity inventory, adjusting commodity sales strategies, increasing investment in tourism facilities and strengthening safety management; commodity sales strategies include but are not limited to pushing coupons, precise delivery and recommending personalized advertisements; The data visualization subsystem presents the customer flow data in the form of chart reports, making the data more intuitive, easy to understand and analyze, and displaying it in real time on an electronic billboard or LCD screen, so that managers can intuitively understand the customer flow situation; chart reports include but are not limited to daily / weekly / monthly customer flow reports, customer behavior analysis reports, customer flow forecast analysis reports, holiday customer flow forecast reports, seasonal customer flow forecast reports, customer portrait and segmentation reports, store operation optimization reports, staffing optimization reports and comprehensive evaluation and improvement suggestion reports.

8. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 7, characterized in that: The store marketing decision subsystem provides accurate customer flow data and analysis reports to support the formulation and implementation of marketing decisions. Based on customer flow data and customer behavior analysis, it evaluates the attractiveness and operating efficiency of the store, conducts comparative analysis with similar stores, identifies gaps and areas for improvement, and makes store optimization suggestions. The optimization suggestions include but are not limited to adjusting shelf positions, optimizing store merchandise display and layout, and optimizing traffic flow design to improve customer shopping experience and purchase conversion rate. The leasing decision subsystem provides data support for the mall's investment promotion and leasing departments, and reasonably determines the store's rental level and lease term based on customer flow and customer characteristic information in different areas. It evaluates the potential value and risks of new stores to provide a reference for investment promotion decisions. It provides a reference for commercial site selection. For companies that are opening new stores or expanding their businesses, it analyzes the customer flow and its characteristics around potential sites to predict the commercial potential of the location and the business formats suitable for operation.

9. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 8, characterized in that: The anomaly detection subsystem identifies those high-flow or low-flow points that deviate from the norm. For the detected anomalies, the system will conduct an in-depth analysis of the relevant factors. Based on the anomaly analysis results, the system can adjust the prediction model. If there are continued anomalies, the system may issue an alarm to the mall manager so that countermeasures can be prepared in advance; abnormal events include but are not limited to temporary events, seasonal changes and system failures; the simulation training subsystem sets up a number of simulation robots to simulate real gender, age, occupation and family status, and conducts simulation optimization of mall services. It simulates all mall service needs of the simulation robots, sets up new categories of products or services in malls and stores, conducts simulation tests, analyzes and learns from the training for multiple times, designs new categories of products or services, analyzes and optimizes according to actual conditions, enters the results into the personal files of the simulation robots, conducts summary statistics and predictive analysis, and finally conducts simulation training on the predictive analysis results and data mining projects.

10. A large-scale shopping mall store customer flow online tracking training prediction analysis system according to claim 9, characterized in that: The mall software subsystem is but not limited to a mall APP, a mall applet or a mall public account; the functions of the mall software subsystem include but are not limited to a 3D positioning navigation map, a navigation route planning module, a related product recommendation module, an interest product recommendation module, an advertising push module, an intelligent route recommendation module and a data collection module; The locations of the 3D positioning navigation map include but are not limited to parking lot locations, parking space locations, store locations, merchandise locations, restroom locations, smoking areas and rest areas; Planning navigation routes: Plan the best 3D navigation route between the user's entry point into the mall and the designated destination, or plan the best 3D navigation route based on the number of products or places the user selects. The route can be changed and reconfirmed at any time according to the user's needs. The best 3D navigation route includes the shortest distance mode, indoor parking mode, outdoor parking mode, lowest price mode and system recommended mode. The related product recommendation module recommends products on the planned 3D navigation route to facilitate users to purchase; The product recommendation module recommends products of interest based on the user's purchase history data, improving the customer's shopping experience and purchase conversion rate; The advertising push module pushes personalized promotional information and coupons based on the customer's shopping history and behavior data; The intelligent route recommendation module intelligently recommends several routes of interest based on the user's shopping history and navigation route history, including but not limited to food routes, shopping routes or overlapping routes; The data collection module will use all the information collected by the mall software subsystem and transmit it to the data analysis layer for big data analysis to obtain more accurate data, enter it into personal files, and form the most accurate personalized service.

Images