Physical chain store management system based on Internet of Things

Abstract

The invention discloses an entity chain store management system based on the Internet of Things, and the system comprises an Internet of Things sensing module, an intelligent analysis and decision support module, a dynamic resource scheduling module, a customer service optimization module, a safety monitoring and emergency response module, and an innovation marketing module. The Internet of Things sensing module is used for comprehensively sensing data of temperature, humidity and illumination in a chain store, positions and stock of commodities and browsing tracks and purchase records of customers. According to the entity chain store management system based on the Internet of Things, the dynamic resource scheduling module captures subtle changes of the stock quantity, the personnel state and the equipment parameters of each chain store in real time through high-frequency data acquisition once every five minutes, the limitation of data lag in traditional static scheduling is broken through, a virtual mapping model is constructed based on 1000 resource nodes, and the real-time performance of the system is improved. A local resource state of a single store is converted into a global visual digital resource network, so that a scheduling decision can cover resource pools of all chain stores.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology and chain store management, specifically to an IoT-based physical chain store management system. Background Technology

[0002] A physical chain store refers to a large number of small, scattered retail stores under the same brand that sell similar goods and services. Under the organization and leadership of the headquarters, they adopt common business policies and consistent marketing actions, and implement an organic combination of centralized procurement and decentralized sales to achieve economies of scale through standardized operations.

[0003] Traditional brick-and-mortar chain store management is characterized by static and localized resource allocation. Inventory distribution, staffing, and equipment usage cannot be dynamically adjusted based on real-time operational data, which can easily lead to resource waste or insufficient supply in scenarios of sales fluctuations and manpower shortages. Summary of the Invention

[0004] The purpose of this invention is to provide an Internet of Things-based physical chain store management system to solve the problems of static and localized resource scheduling in traditional physical chain store management as mentioned in the above-mentioned patents and the inability to dynamically adjust inventory distribution, staffing and equipment usage based on real-time operational data, which can easily lead to resource waste or insufficient supply in scenarios of sales fluctuations and manpower shortages.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a physical chain store management system based on the Internet of Things (IoT), comprising an IoT sensing module, an intelligent analysis and decision support module, a dynamic resource scheduling module, a customer service optimization module, a security monitoring and emergency response module, and an innovative marketing module. The IoT sensing module is responsible for comprehensively sensing the temperature, humidity, and light in the chain store, the location and inventory of goods, and the browsing trajectory and purchase records of customers, and sending the data to the server for storage via the IoT. The intelligent analysis and decision support module is used to perform in-depth mining and analysis of the data collected by the perception module, identify sales hotspots, predict inventory demand, assess customer preferences, and optimize product display. The dynamic resource scheduling module is used to dynamically adjust the inventory distribution, staffing, and equipment usage within the chain stores, specifically: Every 5 minutes, the inventory quantity, staff on-duty status, and equipment operating parameters of each chain store are collected. The inventory quantity is accurate to the unit. The staff on-duty status is divided into three categories: on-duty, off-duty, and busy. The equipment operating parameters include running time and energy consumption value. A virtual mapping model containing 1,000 resource nodes is constructed based on the collected data. The node position error is controlled within 0.5 meters, and the model is updated every 10 minutes. Set constraints such as an inventory turnover rate of no less than 90%, a personnel utilization rate of no less than 85%, and equipment energy consumption of no more than 15% of the rated value. Based on the virtual model and constraint parameters, an initial resource scheduling plan is generated within 20 seconds, including the quantity of inventory to be transferred, personnel job assignments, and equipment start-up and shutdown times. The decision model is trained based on 30 days of historical scheduling data with a 24-hour cycle, and the resource allocation correction coefficient is output. The correction coefficient ranges from 0.8 to 1.2. The initial scheme is optimized using a mixed integer programming algorithm, which solves 5,000 variables within 30 seconds to determine the optimal resource allocation scheme. The system monitors the execution of the plan in real time. When the deviation between the actual value and the plan value exceeds 5%, the plan is refactored. The refactoring process is completed within 40 seconds. The customer service optimization module is used to combine customer behavior data and preference analysis to provide customers with personalized recommendations, fast checkout, and exclusive member benefits. The security monitoring and emergency response module is used to monitor the security status of chain stores in real time and provide emergency plan management functions. The innovative marketing module is used to collect customer behavior data and combine it with market trend analysis to provide innovative marketing strategies for chain stores.

[0006] Preferably, the IoT sensing module uses an RFID reader, a smart sensor, and a high-definition camera, combined with edge computing technology, to achieve real-time data acquisition, preprocessing, and low-latency transmission. The RFID reader has a recognition distance of 3-8 meters, the smart sensor has a sampling frequency of 10Hz, and the high-definition camera has a frame rate of 25fps.

[0007] Preferably, the intelligent analysis and decision support module utilizes advanced technologies such as big data processing, machine learning, and knowledge graphs to construct core analysis models for sales hotspot identification, inventory demand forecasting, customer preference modeling, and product display optimization. The module supports both real-time and offline analysis modes and provides a visual decision dashboard and automated strategy recommendations.

[0008] Preferably, the virtual mapping model of the dynamic resource scheduling module includes a three-dimensional spatial coordinate system, with an X-axis range of 0-50 meters, a Y-axis range of 0-30 meters, and a Z-axis range of 0-5 meters. The resource nodes include three types of identification information: product SKU code, employee number, and equipment number.

[0009] Preferably, the customer service optimization module constructs an omnichannel customer behavior map and a dynamic preference prediction engine, integrates multimodal data fusion, reinforcement learning recommendation and edge computing real-time response technologies, provides personalized services on APP, self-service terminals and smart shelves, and establishes a quantitative evaluation system for service utility.

[0010] Preferably, the safety monitoring and emergency response module includes abnormal behavior recognition, safety hazard detection, emergency plan storage and automatic alarm functions. The abnormal behavior recognition response time is 10 seconds, the safety hazard detection covers three scenarios: fire-fighting facilities, electrical equipment and unobstructed passages, and the emergency plan storage capacity is 50 sets.

[0011] Preferably, the innovative marketing module includes functions for marketing campaign planning, campaign performance monitoring, social media promotion, and coupon management. The marketing campaign planning supports three cycle settings: 7 days, 15 days, and 30 days. The campaign performance monitoring includes three indicators: number of participants, conversion rate, and sales revenue. The coupon management supports full-process control over generation, distribution, and redemption.

[0012] Compared with the prior art, the beneficial effects of the present invention are: This IoT-based physical chain store management system features a dynamic resource scheduling module that captures subtle changes in inventory levels, staff status, and equipment parameters in each chain store in real time through high-frequency data collection every 5 minutes. This overcomes the limitations of data lag in traditional static scheduling. Based on a virtual mapping model built with 1000 resource nodes (node ​​position error ≤ 0.5 meters), it transforms the local resource status of a single store into a globally visible digital resource network. This allows scheduling decisions to cover the resource pools of all chain stores, avoiding resource imbalances between regions caused by traditional local scheduling. By setting rigid constraints such as inventory turnover rate ≥ 90%, staff utilization rate ≥ 85%, and equipment energy consumption ≤ 15% of rated value, combined with a mixed integer programming algorithm, it completes the calculation of 5000 variables within 30 seconds. This allows the entire process of resource allocation scheme generation and optimization to be controlled within 50 seconds, improving efficiency by more than 80% compared to traditional manual scheduling. For scenarios with sales fluctuations, when the deviation between actual and planned values ​​exceeds 5%, the scheme can be reconstructed within 40 seconds. To ensure real-time matching of resource supply and demand changes, a periodic model based on 30 days of historical data is trained 24 hours / cycle, outputting a correction coefficient of 0.8-1.2. This allows the scheduling scheme to dynamically adapt to operational patterns at different times. For example, during peak customer traffic periods, dynamic allocation of personnel positions maintains staff utilization at over 85%, avoiding idle manpower. In inventory management, precise allocation stabilizes turnover at over 90%, reducing the double losses of slow-moving inventory and stockouts of best-selling items. Optimized control of equipment start-up and shutdown periods ensures energy consumption does not exceed 15% of the rated value, reducing ineffective energy expenditure. Through the combination of a virtual mapping model and a real-time monitoring mechanism, in the event of sudden equipment failures or manpower shortages, the system can quickly identify resource gaps and initiate cross-store collaborative scheduling. For example, when a store experiences a decrease in production capacity due to equipment failure, the system can generate allocation instructions for idle equipment in surrounding stores within one minute based on the equipment redundancy status of each store in the virtual model, avoiding the response delays caused by information asymmetry in traditional scheduling. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the safety monitoring and emergency response module of the present invention; Figure 2 This is a schematic diagram of the innovative marketing module of the present invention; Figure 3 This is a schematic diagram illustrating the principle of the Internet of Things-based physical chain store management system of the present invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figure 1-3 This invention provides a technical solution: an Internet of Things (IoT)-based management system for physical chain stores, comprising an IoT sensing module, an intelligent analysis and decision support module, a dynamic resource scheduling module, a customer service optimization module, a security monitoring and emergency response module, and an innovative marketing module. The IoT sensing module is responsible for comprehensively sensing the temperature, humidity, and light intensity within the chain store, as well as the location and inventory of goods, and the browsing trajectory and purchase records of customers. The data is then transmitted to the server for storage via the IoT. The temperature and humidity sensing range covers -10℃ to 50℃ and 10%RH to 90%RH. The light intensity acquisition supports automatic range switching. The goods location is achieved through dual verification using RFID and infrared positioning. Inventory data acquisition is triggered synchronously when goods enter or leave the warehouse. Customer browsing trajectory records start at the store entrance. Purchase records include details of goods, quantity, unit price, and payment method.

[0016] The intelligent analysis and decision support module is used to deeply mine and analyze the data collected by the perception module, identify sales hotspots, predict inventory demand, assess customer preferences, and optimize product display. Sales hotspot identification calculates sales per unit time based on shelf zones and generates a ranking of hotspot areas. Inventory demand prediction models the data separately for weekdays, weekends, and holidays, and outputs suggested replenishment quantities for each time period. Customer preference assessment generates preference levels by analyzing purchase frequency, intervals between product categories, and sales amounts. Product display optimization provides shelf location adjustment plans based on product relevance and sales volume.

[0017] The dynamic resource scheduling module is used to dynamically adjust the inventory distribution, staffing, and equipment usage within the chain stores, specifically: Every 5 minutes, the inventory quantity, staff on-duty status, and equipment operating parameters of each chain store are collected. The inventory quantity is counted separately according to the shelf location. The staff on-duty status is determined by a combination of the attendance system and mobile terminal positioning. The equipment operating parameters also include operating temperature and fault warning information. A virtual mapping model containing 1,000 resource nodes is constructed based on the collected data. The node positions are marked according to the actual coordinates, and the node correlation is checked synchronously when the model is updated. Set constraints such as an inventory turnover rate of no less than 90%, a personnel utilization rate of no less than 85%, and equipment energy consumption of no more than 15% of the rated value. All parameters are statistically summarized daily, weekly, and monthly for analysis. Based on the virtual model and constraint parameters, an initial resource scheduling plan is generated within 20 seconds. The inventory allocation quantity takes into account the loss during transportation, the personnel job assignment matches the personnel skill level, and the equipment start-up and shutdown times avoid peak passenger flow periods. The decision-making model is trained based on 30 days of historical scheduling data with a 24-hour cycle, and the resource allocation correction coefficient is output. The correction coefficient is dynamically fine-tuned according to real-time operation data. A mixed-integer programming algorithm is used to optimize the initial solution, completing the calculation of 5000 variables within 30 seconds to determine the optimal resource allocation scheme. During optimization, various constraint parameters are balanced. The implementation of the scheme is monitored in real time. When the deviation between the actual value and the scheme value exceeds 5%, a scheme reconstruction is triggered, completed within 40 seconds. During reconstruction, priority is given to ensuring the supply of core goods and the allocation of key personnel. The customer service optimization module combines customer behavior data and preference analysis to provide customers with personalized recommendations, fast checkout, and exclusive member discounts. Personalized recommendations are generated based on the three most recent purchase records and real-time browsing behavior. Fast checkout supports multiple payment methods in parallel processing, and exclusive member discounts are set with different discount levels based on membership level and purchase amount.

[0018] The security monitoring and emergency response module is used to monitor the security status of the chain stores in real time and provide emergency plan management functions; the security status monitoring includes areas with abnormal personnel density and movement, the emergency response is handled according to the severity of the event, and the plan execution process clearly defines the responsible persons and contact information for each link.

[0019] The innovative marketing module collects customer behavior data and, combined with market trend analysis, provides innovative marketing strategies for chain stores. Customer behavior data is categorized and summarized daily, market trend analysis tracks industry dynamics and changes in consumer trends, and marketing strategies include event themes, times, locations, content, and promotional methods. The IoT sensing module uses RFID readers, smart sensors, and high-definition cameras, combined with edge computing technology, to achieve real-time data acquisition, preprocessing, and low-latency transmission. RFID readers are installed at shelf entrances and exits and at the cash register, smart sensors are distributed throughout the store and on various devices, and high-definition cameras cover shelf aisles, cash registers, and the inside and outside of the store.

[0020] The intelligent analysis and decision support module utilizes advanced technologies such as big data processing, machine learning, and knowledge graphs to construct core analysis models for sales hotspot identification, inventory demand forecasting, customer preference modeling, and product display optimization. The module analyzes real-time sales data, processes historical accumulated data offline, displays key indicator change curves on a visual decision dashboard, and recommends automated strategies in priority order.

[0021] The virtual mapping model of the dynamic resource scheduling module includes a three-dimensional spatial coordinate system: X-axis range 0-50 meters, Y-axis range 0-30 meters, and Z-axis range 0-5 meters. Resource nodes contain unique product information corresponding to product SKU codes, employee IDs associated with basic personnel information and skills, and equipment numbers corresponding to equipment model parameters and maintenance records. The relationships between nodes reflect the direction of resource flow and dependencies. The customer service optimization module constructs an omnichannel customer behavior map and a dynamic preference prediction engine, integrates multimodal data to process data from different sources, uses reinforcement learning to continuously adjust strategies based on feedback, edge computing provides real-time response to ensure timely service, displays personalized recommendations in the APP, supports quick operation on self-service terminals, displays product discount information on smart shelves, and a service utility quantification evaluation system generates evaluation reports regularly.

[0022] The security monitoring and emergency response module includes abnormal behavior identification to mark suspicious behaviors, security hazard detection to generate detection reports regularly, emergency plan storage to be classified and organized by event type, automatic alarm to be activated immediately when the trigger conditions are met, abnormal behavior identification to cover theft and fighting, security hazard detection to regularly check the status of facilities, emergency plans to include response steps for different scenarios, and automatic alarm to notify relevant personnel and security departments.

[0023] The innovative marketing module includes marketing campaign planning to develop plans based on holidays and peak sales seasons, monitoring and tracking changes in various indicators to track campaign effectiveness, social media promotion to select appropriate platforms to publish information, coupon management to control the number of coupons issued and their scope of use, marketing campaign planning to clarify campaign rules and participation methods, campaign effectiveness monitoring to compare sales data before and after the campaign, social media promotion to publish content according to plan, and coupon management to record usage and effectiveness.

[0024] Example 1: A chain supermarket This supermarket chain has deployed IoT sensing modules in each store. RFID readers are installed at shelf entrances and checkout counters. Smart sensors are distributed throughout the store, including refrigerated display cases and air conditioners. High-definition cameras cover shelf aisles, checkout counters, and the inside and outside of the store. Temperature and humidity sensors monitor the store environment in real time, ranging from -10℃ to 50℃ and 10%RH to 90%RH. When the temperature inside a refrigerated display case exceeds the set range, the data is collected and sent to the server. Light intensity monitoring supports automatic range switching, adjusting the accuracy of the store's lighting based on changes in external light. When goods are put on shelves, their location is determined through dual verification using RFID and infrared positioning. Inventory data is collected synchronously when goods enter or leave the warehouse, ensuring accurate inventory levels. Customers' browsing trajectory is recorded from the store entrance, every 0.Sampling occurs every 5 seconds. Purchase records include product details, quantity, unit price, and payment method. This data is transmitted to a server for storage via the Internet of Things (IoT). The intelligent analysis and decision support module processes the collected data, calculating sales per unit time by shelf area, generating rankings of hotspot areas, and identifying fresh produce and snack areas as sales hotspots. Inventory demand is predicted based on weekdays, weekends, and holidays; for example, it predicts that replenishment of the fresh produce area needs to be 30% higher on weekends than on weekdays. By analyzing customer purchase frequency, intervals, categories, and amounts, preference levels are generated, revealing that younger customers prefer imported snacks. Based on product relevance and sales volume, shelf-specific recommendations are provided. The location adjustment plan places yogurt and bread on adjacent shelves. This module supports both real-time and offline analysis modes. Real-time analysis analyzes immediate sales data, while offline analysis analyzes historical cumulative data. A visual decision dashboard displays key indicator curves for sales and customer traffic. Automated strategy recommendations are prioritized for management reference. The dynamic resource scheduling module collects inventory quantity, staff attendance status, and equipment operating parameters for each store every 5 minutes. Inventory quantity is calculated by shelf location, such as 20 bags of a certain brand of potato chips remaining on shelf A in the snack area. Staff attendance status is determined through a combination of the attendance system and mobile terminal location, distinguishing between three states: on-duty, off-duty, and busy. The busy state... The status is judged based on a customer reception time of ≥3 minutes; equipment operating parameters include operating time, energy consumption, operating temperature, and fault warning information, such as the refrigerated display case operating temperature of 4℃ and energy consumption of 2kWh / h. Based on this data, a virtual mapping model containing 1000 resource nodes is constructed. The node positions are marked in a three-dimensional spatial coordinate system according to actual coordinates (X-axis 0-50 meters, Y-axis 0-30 meters, Z-axis 0-5 meters). Resource nodes include product SKU codes, employee numbers, and equipment numbers, corresponding to unique product information, basic personnel information and skill information, equipment model parameters, and maintenance records, respectively. The model is updated every 10 minutes, and node associations are verified synchronously. The system sets constraints such as an inventory turnover rate of no less than 90%, a personnel utilization rate of no less than 85%, and equipment energy consumption not exceeding 15% of the rated value. These parameters are statistically analyzed daily, weekly, and monthly. Based on the virtual model and constraints, an initial resource scheduling plan is generated within 20 seconds. Considering losses during transportation, 50 bags of a certain brand of potato chips are allocated from the warehouse to the store. Personnel assignments are matched to skill levels, with employees experienced in fresh produce handling assigned to the fresh produce area. Equipment start-up and shutdown times are avoided during peak customer traffic periods; for example, air conditioning temperatures are appropriately increased during off-peak hours to save energy. The decision model is trained based on 30 days of historical scheduling data, using a 24-hour cycle, and outputs a value of 0.8-1.A resource allocation correction coefficient of 2 is applied and dynamically fine-tuned based on real-time operational data. A mixed-integer programming algorithm is used to solve for 5000 variables within 30 seconds, optimizing the initial plan, balancing various constraint parameters, determining the optimal resource allocation plan, and monitoring the plan's execution in real time. When the actual inventory deviates from the plan value by more than 5%, a plan reconstruction is triggered, completed within 40 seconds. During reconstruction, priority is given to ensuring the supply of core fresh produce and the allocation of key personnel at the checkout counter. The customer service optimization module provides services based on customer behavior data and preference analysis, utilizing young customers' most recent three purchase records and real-time browsing behavior of imported snacks. The app provides personalized recommendations for new imported snacks; quick checkout supports multiple payment methods, including facial recognition and QR code scanning, shortening checkout time; exclusive member discounts are set according to membership level and spending amount, with gold card members enjoying a 20% discount on purchases over 200 yuan; this module constructs a full-channel customer behavior map and dynamic preference prediction engine, integrating multimodal data, using reinforcement learning recommendations to continuously adjust based on customer feedback, edge computing to ensure timely service response, self-service terminals to support fast operation, smart shelves to display product discount information, a service utility quantitative evaluation system to regularly generate evaluation reports, and security monitoring and response. The emergency response module monitors the store's security in real time, using high-definition cameras to identify crowd density and areas of abnormal movement. If excessive crowding is detected in the fresh produce area at a certain time, an alert is issued promptly. Abnormal behavior identification covers theft and fighting, marking suspicious activities with a 10-second response time. The safety hazard detection system regularly checks fire safety facilities, electrical equipment, and passageway accessibility, generating inspection reports. Emergency plans are stored and organized into 50 sets, categorized by fire and power outage types, with clear instructions on responsible personnel and contact information for each step of the plan's execution. When a safety hazard or abnormal behavior is detected, an automatic alarm is immediately activated, notifying the relevant personnel. The Human Resources and Security departments, through an innovative marketing module, collected customer behavior data, categorized and summarized it daily, and combined it with market trend analysis to track the rise of health food consumption. They then planned a "Healthy Living Week" marketing campaign for the stores. The campaign took place on weekends in-store and included health food tastings and nutritionist lectures. Information about the campaign was disseminated through social media platforms, and electronic coupons were created, with controlled distribution and usage scope, valid only for purchases of health foods during the campaign period. The campaign's effectiveness was monitored by tracking participation, conversion rates, and sales figures, comparing sales data before and after the campaign, evaluating its impact, and managing coupon usage and results.

[0025] In summary, this IoT-based physical chain store management system utilizes a dynamic resource scheduling module that captures subtle changes in inventory levels, staff status, and equipment parameters in each chain store in real time through high-frequency data collection every 5 minutes. This overcomes the limitations of data lag in traditional static scheduling. Based on a virtual mapping model constructed with 1000 resource nodes (node ​​position error ≤ 0.5 meters), it transforms the local resource status of a single store into a globally visible digital resource network. This allows scheduling decisions to cover the resource pools of all chain stores, avoiding resource imbalances between regions caused by traditional local scheduling. By setting rigid constraints such as inventory turnover rate ≥ 90%, staff utilization rate ≥ 85%, and equipment energy consumption ≤ 15% of rated value, combined with a mixed integer programming algorithm, it completes the calculation of 5000 variables within 30 seconds. This allows the entire process of resource allocation scheme generation and optimization to be controlled within 50 seconds, improving efficiency by more than 80% compared to traditional manual scheduling. For scenarios with sales fluctuations, when the deviation between actual and planned values ​​exceeds 5%, a solution can be completed within 40 seconds. Reconstruction ensures real-time matching of resource supply and demand changes. A periodic model, trained 24 hours / cycle based on 30 days of historical data, outputs a correction coefficient of 0.8-1.2, enabling the scheduling scheme to dynamically adapt to operational patterns at different times. For example, during peak customer traffic periods, dynamic allocation of personnel positions maintains staff utilization above 85%, avoiding idle manpower. In inventory management, precise allocation stabilizes turnover above 90%, reducing the double losses of slow-moving inventory and stockouts of best-selling items. Optimized control of equipment start-up and shutdown periods ensures energy consumption does not exceed 15% of the rated value, reducing ineffective energy expenditure. Through the combination of a virtual mapping model and a real-time monitoring mechanism, in the event of equipment failure or manpower shortages, the system can quickly identify resource gaps and initiate cross-store collaborative scheduling. For instance, when a store experiences a decrease in production capacity due to equipment failure, the system can generate allocation instructions for idle equipment in surrounding stores within one minute based on the equipment redundancy status of each store in the virtual model, avoiding the response delays caused by information asymmetry in traditional scheduling.

[0026] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0027] The relevant modules involved in this system are all hardware system modules or functional modules that combine computer software programs or protocols with hardware in the prior art. The computer software programs or protocols involved in these functional modules are technologies known to those skilled in the art and are not improvements to this system. The improvement of this system lies in the interaction or connection between the modules, that is, in improving the overall structure of the system to solve the corresponding technical problems that this system aims to address.

[0028] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A physical chain store management system based on the Internet of Things (IoT), comprising an IoT sensing module, an intelligent analysis and decision support module, a dynamic resource scheduling module, a customer service optimization module, a security monitoring and emergency response module, and an innovative marketing module, characterized in that: The IoT sensing module is responsible for comprehensively sensing the temperature, humidity, and light in the chain stores, the location and inventory of goods, as well as the browsing trajectory and purchase records of customers, and sending the data to the server for storage via the Internet of Things. The intelligent analysis and decision support module is used to perform in-depth mining and analysis of the data collected by the perception module, identify sales hotspots, predict inventory demand, assess customer preferences, and optimize product display. The dynamic resource scheduling module is used to dynamically adjust the inventory distribution, staffing, and equipment usage within the chain stores, specifically: Every 5 minutes, the inventory quantity, staff on-duty status, and equipment operating parameters of each chain store are collected. The inventory quantity is accurate to the unit. The staff on-duty status is divided into three categories: on-duty, off-duty, and busy. The equipment operating parameters include running time and energy consumption value. A virtual mapping model containing 1,000 resource nodes is constructed based on the collected data. The node position error is controlled within 0.5 meters, and the model is updated every 10 minutes. Set constraints such as an inventory turnover rate of no less than 90%, a personnel utilization rate of no less than 85%, and equipment energy consumption of no more than 15% of the rated value. Based on the virtual model and constraint parameters, an initial resource scheduling plan is generated within 20 seconds, including the quantity of inventory to be transferred, personnel job assignments, and equipment start-up and shutdown times. The decision model is trained based on 30 days of historical scheduling data with a 24-hour cycle, and the resource allocation correction coefficient is output. The correction coefficient ranges from 0.8 to 1.

2. The initial scheme is optimized using a mixed integer programming algorithm, which solves 5,000 variables within 30 seconds to determine the optimal resource allocation scheme. The system monitors the execution of the plan in real time. When the deviation between the actual value and the plan value exceeds 5%, the plan is refactored. The refactoring process is completed within 40 seconds. The customer service optimization module is used to combine customer behavior data and preference analysis to provide customers with personalized recommendations, fast checkout, and exclusive member benefits. The security monitoring and emergency response module is used to monitor the security status of chain stores in real time and provide emergency plan management functions. The innovative marketing module is used to collect customer behavior data and combine it with market trend analysis to provide innovative marketing strategies for chain stores.

2. The physical chain store management system based on the Internet of Things according to claim 1, characterized in that: The IoT sensing module uses RFID readers, smart sensors, and high-definition cameras, combined with edge computing technology, to achieve real-time data acquisition, preprocessing, and low-latency transmission. The RFID reader has a recognition distance of 3-8 meters, the smart sensor has a sampling frequency of 10Hz, and the high-definition camera has a frame rate of 25fps.

3. The physical chain store management system based on the Internet of Things according to claim 1, characterized in that: The intelligent analysis and decision support module utilizes advanced technologies such as big data processing, machine learning, and knowledge graphs to construct core analysis models for sales hotspot identification, inventory demand forecasting, customer preference modeling, and product display optimization. The module supports both real-time and offline analysis modes and provides a visual decision dashboard and automated strategy recommendations.

4. The physical chain store management system based on the Internet of Things according to claim 1, characterized in that: The virtual mapping model of the dynamic resource scheduling module includes a three-dimensional spatial coordinate system, with the X-axis ranging from 0 to 50 meters, the Y-axis ranging from 0 to 30 meters, and the Z-axis ranging from 0 to 5 meters. Resource nodes include three types of identification information: product SKU code, employee number, and equipment number.

5. The physical chain store management system based on the Internet of Things according to claim 1, characterized in that: The customer service optimization module constructs an omnichannel customer behavior map and a dynamic preference prediction engine, integrating multimodal data fusion, reinforcement learning recommendation, and edge computing real-time response technologies to provide personalized services through apps, self-service terminals, and smart shelves, and establishes a quantitative evaluation system for service effectiveness.

6. The physical chain store management system based on the Internet of Things according to claim 1, characterized in that: The safety monitoring and emergency response module includes abnormal behavior recognition, safety hazard detection, emergency plan storage and automatic alarm functions. The abnormal behavior recognition response time is 10 seconds. The safety hazard detection covers three scenarios: fire-fighting facilities, electrical equipment and unobstructed passages. The emergency plan storage capacity is 50 sets.

7. The physical chain store management system based on the Internet of Things according to claim 1, characterized in that: The innovative marketing module includes functions for marketing campaign planning, campaign performance monitoring, social media promotion, and coupon management. Marketing campaign planning supports three cycle settings: 7 days, 15 days, and 30 days. Campaign performance monitoring includes three indicators: number of participants, conversion rate, and sales. Coupon management supports full-process control of generation, issuance, and redemption.

Images