Station management system based on Internet of Things

Abstract

The invention discloses a station management system based on the Internet of Things, and the system comprises an Internet of Things sensing module which comprises an RFID tag, a temperature and humidity sensor, a weight sensor, a position sensor and a camera, the RFID tag recognizes a container, the temperature and humidity sensor, the weight sensor and the position sensor collect cargo information, and the camera collects image information in a station; the beneficial effects of the system are that comprehensive integration and intelligent management of on-site management, warehouse management, container operation, statistical analysis and other functions of the comprehensive taxation area station are realized, and the efficiency, the accuracy and the safety of station management are improved through cooperative work among the modules; by means of the Internet of Things sensing module, comprehensive sensing and real-time monitoring of various devices, goods, transportation tools and the like in the station are achieved, accurate and timely information support is provided for station management, and timeliness and accuracy of management are effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of field management technology, specifically relating to a field management system based on the Internet of Things. Background Technology

[0002] Comprehensive bonded zone management is the process by which customs comprehensively supervises and manages the facilities, goods, means of transport, and enterprises within the comprehensive bonded zone. It aims to ensure that the operation of the zone complies with customs supervision requirements, safeguard trade security and facilitation, and promote high-quality regional economic development.

[0003] A site management method and system based on video fusion, disclosed in patent publication number CN119992402A, includes: a 3D scene model generation module, which performs 3D modeling of the target site based on 3D measurement information to generate a 3D scene model of the target site; a multiple video source determination module, which traverses multiple video acquisition points within the target site to determine multiple video sources; a panoramic video data construction module, which performs real-time stitching based on the multiple video sources to construct panoramic video data of the target site; a 3D scene fusion model generation module, which uses the panoramic video data mapped onto the 3D scene model for fusion to generate a 3D scene fusion model; an anomaly information set extraction module, which performs 3D inspection of the target site using the 3D scene fusion model and extracts anomaly information set based on the inspection results; and an alarm command generation module, which locates faults in the target site according to the anomaly information set and generates alarm commands based on the fault location information.

[0004] The system mainly focuses on visual monitoring and anomaly management of the site through video fusion technology. It lacks comprehensive and in-depth integration and interaction of the complex and diverse equipment and information within the site, making it difficult to achieve refined and intelligent collaborative management of all aspects of site operation. It can only trigger alarms after an anomaly occurs, and cannot achieve proactive optimization of work processes and risk prediction. Summary of the Invention

[0005] The purpose of this invention is to provide an IoT-based terminal management system that integrates and intelligently manages functions such as on-site management, warehouse management, container operations, and statistical analysis of comprehensive bonded zone terminals, thereby improving the efficiency, accuracy, and security of terminal management; enabling proactive optimization of work processes and risk prediction; achieving comprehensive perception and real-time monitoring through IoT sensing devices; and utilizing big data analytics and machine learning algorithms to achieve intelligent management and decision support.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a site management system based on the Internet of Things, comprising...

[0007] The Internet of Things (IoT) sensing module includes RFID tags, temperature and humidity sensors, weight sensors, position sensors, and cameras. The RFID tags identify containers, the temperature and humidity sensors, weight sensors, and position sensors collect cargo information, and the cameras collect image information within the terminal.

[0008] The data processing module cleans, classifies, and stores the collected information for preprocessing. It uses big data analytics and machine learning algorithms to analyze and mine information on cargo flow patterns, equipment failure modes, and transportation efficiency.

[0009] The on-site management module obtains container arrival information through the IoT sensing module to update the on-site status, verifies container departure information to update system records, and tracks vehicle entry, exit, parking and cargo loading information in real time to optimize scheduling.

[0010] The warehouse management module plans storage areas based on the type, attributes, and storage requirements of goods, monitors the storage status of goods through the Internet of Things sensing module, and optimizes the sorting path of goods based on order information and data analysis results.

[0011] The container operation management module uses optimization algorithms to generate operation scheduling plans based on the operation tasks and the actual situation of the terminal, arranges operation equipment and personnel to perform container unloading, container lifting and container moving operations, and monitors the operation process in real time;

[0012] The statistical analysis module performs statistical analysis on indicators such as cargo throughput, equipment utilization, operational efficiency, and operating costs, presents the results in the form of charts and reports, and predicts the future operation of the station based on historical data.

[0013] The risk prediction module uses machine learning algorithms to build a risk prediction model, predict the risks of cargo damage, equipment failure, and transportation delays, generate risk warning information, and send it to management personnel and the decision support module.

[0014] The decision support module receives risk warning information, decision suggestions, and statistical analysis results, and provides decision-making solutions based on the actual operation and management objectives of the site, supporting decision simulation and evaluation.

[0015] The communication module adopts wireless network communication protocol and Bluetooth communication protocol to realize stable data transmission and information sharing between modules and between remote control terminal and customs supervision system;

[0016] The user interaction module supports managers in querying data, issuing operation commands, and setting system parameters. It also displays the real-time operating status of the site, statistical analysis results, and risk warning information, and receives feedback from managers to optimize system functions and user experience.

[0017] The system management module manages system user permissions, data security, and system logs.

[0018] As a preferred technical solution of the present invention, the data processing module is implemented as follows:

[0019] After receiving the raw data transmitted by the IoT sensing module, the system performs data cleaning; by setting data format rules, it removes or corrects data that does not meet the standards; it uses data filling algorithms to supplement missing data; and it stores the data in a distributed database according to data type.

[0020] Big data analytics is used to perform in-depth processing on categorized and stored data; the MapReduce framework is used for parallel computing of massive amounts of data; clustering algorithms are used to analyze the driving trajectory and loading status of transportation vehicles to calculate the utilization efficiency of transportation vehicles; and the valuable information obtained from the analysis and mining is organized into decision suggestions and transmitted to the decision support module in the form of structured data.

[0021] As a preferred technical solution of the present invention, the data filling algorithm includes mean filling and linear interpolation.

[0022] As a preferred technical solution of the present invention, the on-site management module is implemented as follows:

[0023] When a container enters the terminal area, an RFID reader installed at the terminal entrance reads the RFID tag information on the container, obtains the container number, cargo type, and arrival time information, and transmits this information to the on-site management module. The on-site management module updates the container's on-site status to "arrived" in the system based on the received information, and records the relevant information to the database. When the container leaves, the RFID tag is read again at the exit to verify whether the departure information matches the system record. If they match, the container's on-site status is updated to "departed," and the departure time information is recorded.

[0024] License plate recognition cameras are installed at the entrances and exits of the depot and in the corresponding areas to collect vehicle entry and exit information in real time, including license plate number and entry and exit time. By connecting with the transportation management system, information on the cargo loaded on the vehicles is obtained. The on-site management module uses geographic information system technology to track and display the driving trajectory and parking location of vehicles in the depot in real time. Based on historical vehicle scheduling data and the current depot operation status, a genetic algorithm is used to calculate the optimal vehicle scheduling plan.

[0025] As a preferred technical solution of the present invention, the warehouse management module is implemented as follows:

[0026] Before goods are put into storage, the warehouse management module uses a space planning algorithm to generate the optimal storage area allocation scheme based on the type, attributes and storage requirements of the goods and the spatial layout of the warehouse. The warehouse is divided into different storage areas and each area is assigned a unique identifier.

[0027] The temperature and humidity sensors and location sensors in the IoT sensing module collect real-time information on the storage status of goods and transmit the data to the warehouse management module. The warehouse management module analyzes the data and issues an alert if any abnormality is detected. In the goods sorting process, the list of goods to be sorted is determined based on the order information. Combining the storage location of the goods and the data analysis results, the path planning algorithm is used to generate the shortest sorting path to guide the sorting personnel to complete the sorting operation.

[0028] As a preferred technical solution of the present invention, the abnormal situation includes excessively high temperature and cargo displacement exceeding the set range.

[0029] As a preferred technical solution of the present invention, the container operation management module is implemented as follows:

[0030] When a new container operation task is received, the container operation management module first obtains the current situation of the terminal, including the usage status of the operation equipment, the work arrangement of the operation personnel, and the location information of the containers on site; then, it uses a mixed integer programming algorithm to generate the optimal operation scheduling plan with the goal of minimizing operation time and maximizing equipment utilization, and determines the operation equipment, operation sequence, and operation time parameters.

[0031] During the operation, sensors and cameras installed on the equipment collect real-time data on the equipment's operating status and the container's location, which is then transmitted to the container operation management module. The module monitors the operation in real time, and if any abnormality is detected, it immediately issues an alarm, suspends the relevant operation, notifies maintenance or management personnel for handling, and records all data during the operation.

[0032] As a preferred technical solution of the present invention, the statistical analysis module is implemented as follows:

[0033] Extract statistical data on cargo throughput, equipment utilization, operational efficiency, and operating costs from the database; and use data visualization technology to present the statistical results in the form of charts and reports.

[0034] Time series analysis and machine learning algorithms are used to train historical data to establish a terminal operation prediction model; based on the trends and patterns of historical data, cargo throughput, equipment demand, and workload indicators are predicted.

[0035] As a preferred technical solution of the present invention, the cargo throughput is calculated by counting the number of goods entering and leaving the site over a time period; the equipment utilization rate is calculated by the ratio of the actual working time of the equipment to the planned working time; the operating efficiency is calculated based on the number of tasks completed per unit time; and the operating cost statistics include equipment maintenance costs, personnel wages, and various energy consumption expenses.

[0036] As a preferred technical solution of the present invention, the risk prediction module is implemented as follows:

[0037] Historical data related to cargo damage, equipment failure, and transportation delays were collected, and machine learning algorithms were used to train the data to construct cargo damage risk prediction models, equipment failure risk prediction models, and transportation delay risk prediction models, respectively. During the training process, the model parameters were optimized using cross-validation.

[0038] The system acquires data transmitted from IoT sensing modules and other modules in real time, inputs the data into the corresponding risk prediction model for calculation, and generates detailed risk warning information when the model predicts a high risk. The warning information is then sent to the terminal devices and decision support modules of management personnel.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] It has achieved comprehensive integration and intelligent management of functions such as on-site management, warehouse management, container operations and statistical analysis of the comprehensive bonded zone. Through the collaborative work between various modules, it has improved the efficiency, accuracy and security of the zone management.

[0041] With the help of IoT sensing modules, comprehensive sensing and real-time monitoring of various equipment, goods, and transportation vehicles within the station were achieved, providing accurate and timely information support for station management and effectively improving the timeliness and accuracy of management.

[0042] By utilizing big data analytics and machine learning algorithms, in-depth analysis and mining of the collected data are achieved. This not only enables proactive optimization of work processes but also provides early warnings of risks during site operation through a risk prediction module, thereby reducing operational risks. Attached Figure Description

[0043] Figure 1 This is a system schematic diagram of the present invention. Detailed Implementation

[0044] 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.

[0045] Example 1

[0046] Please see Figure 1 This is the first embodiment of the present invention, which provides an Internet of Things (IoT) based site management system, including...

[0047] The Internet of Things (IoT) sensing module includes RFID tags, temperature and humidity sensors, weight sensors, position sensors, and cameras. RFID tags identify containers, temperature and humidity sensors, weight sensors, and position sensors collect cargo information, and cameras collect image information within the site.

[0048] The data processing module cleans, classifies, and stores the collected information for preprocessing. It then uses big data analytics and machine learning algorithms to analyze and extract information on cargo flow patterns, equipment failure modes, and transportation vehicle utilization efficiency. The implementation method of the data processing module is as follows:

[0049] After receiving the raw data transmitted by the IoT sensing module, the system first performs data cleaning. By setting data format rules, data that does not meet the standards is removed or corrected, such as filtering abnormal values ​​collected by the sensors (e.g., temperature and humidity data that are outside the normal range). Data filling algorithms are used to supplement missing data, such as mean filling and linear interpolation methods. Then, the data is classified and stored according to data type (equipment data, cargo data, transportation data) and stored in a distributed database, such as HBase, to improve data storage and query efficiency.

[0050] Big data analytics is used to perform in-depth processing on categorized and stored data; the MapReduce framework is used for parallel computing of massive amounts of data, such as statistically analyzing the quantity and direction of cargo flow over different time periods to identify cargo flow patterns; machine learning algorithms are combined, such as using decision tree algorithms to train equipment operation data, build equipment failure prediction models, and uncover equipment failure modes; clustering algorithms are used to analyze data such as the driving trajectory and loading status of transportation vehicles to calculate the utilization efficiency of transportation vehicles; finally, the valuable information obtained from the analysis and mining is organized into decision recommendations and transmitted to the decision support module in structured data form.

[0051] The on-site management module acquires container arrival information and updates the on-site status through the IoT sensing module, verifies container departure information and updates system records, and tracks vehicle entry, exit, parking, and cargo loading information in real time to optimize scheduling. The implementation method of the on-site management module is as follows:

[0052] When a container enters the terminal area, an RFID reader installed at the terminal entrance reads the RFID tag information on the container, obtaining data such as container number, cargo type, and arrival time, and transmits this information to the on-site management module. The on-site management module updates the container's on-site status to "arrived" based on the received information and records the relevant information in the database. When the container leaves, the RFID tag is read again at the exit to verify whether the departure information (such as cargo quantity, destination, etc.) matches the system record. If they match, the container's on-site status is updated to "departed," and the departure time and other information are recorded.

[0053] License plate recognition cameras are installed at the station entrances and key areas to collect vehicle entry and exit information in real time, including license plate numbers and entry / exit times. By connecting with the transportation management system, information on vehicle cargo loading is obtained. The on-site management module uses geographic information system technology to track and display the driving trajectory and parking location of vehicles within the station in real time. Based on historical vehicle scheduling data and current station operations, optimization algorithms such as genetic algorithms are used to calculate the optimal vehicle scheduling plan, such as rationally arranging vehicle parking positions and loading / unloading sequences, thereby improving station traffic efficiency.

[0054] The warehouse management module plans storage areas based on product type, attributes, and storage requirements, monitors product storage status through an IoT sensing module, and optimizes product sorting routes based on order information and data analysis results. The implementation method of the warehouse management module is as follows:

[0055] Before goods are received into the warehouse, the warehouse management module generates the optimal storage area allocation scheme based on the type of goods (such as electronic products, food, etc.), attributes (fragile goods, dangerous goods, etc.), and storage requirements (temperature, humidity conditions, etc.), combined with the spatial layout of the warehouse, using spatial planning algorithms (such as simulated annealing algorithm); the warehouse is divided into different storage areas and each area is assigned a unique identifier to facilitate the storage and management of goods.

[0056] The IoT sensing module collects real-time information on the storage status of goods, such as temperature and humidity changes and goods movement, through sensors like temperature and humidity sensors and location sensors. This data is then transmitted to the warehouse management module. The warehouse management module analyzes the data and issues timely warnings if any anomalies are detected (such as excessively high temperatures or goods movement exceeding set limits). In the goods sorting process, the module determines the list of goods to be sorted based on order information. Combining the goods storage location with data analysis results (such as the location of frequently shipped goods), it uses path planning algorithms (such as Dijkstra's algorithm) to generate the shortest sorting path, guiding sorting personnel to quickly complete the sorting operation and improving sorting efficiency.

[0057] The container operations management module uses optimization algorithms to generate operation scheduling plans based on the operation tasks and actual conditions at the terminal, arranges operation equipment and personnel to perform container unloading, lifting, and moving operations, and monitors the operation process in real time. The implementation method of the container operations management module is as follows:

[0058] When a new container operation task (such as container unloading, container lifting, or container relocation) is received, the container operation management module first obtains the current situation of the terminal, including the usage status of operating equipment (cranes, forklifts, etc.), the work arrangements of operators, and the location of containers on site. Then, it uses optimization algorithms such as mixed integer programming to generate the optimal operation scheduling plan with the goal of minimizing operation time and maximizing equipment utilization, and determines parameters such as operating equipment, operation sequence, and operation time.

[0059] During operation, sensors (such as position sensors and pressure sensors) and cameras installed on the operating equipment collect real-time information such as the operating status of the equipment and the operating position of the container, and transmit it to the container operation management module. The container operation management module monitors the operation process in real time. If any abnormality is detected (such as equipment failure or operational violation), it immediately issues an alarm, suspends the relevant operation, and notifies maintenance or management personnel for handling, ensuring the safe and efficient operation. At the same time, it records various data during the operation, such as the start time, end time, and equipment operating parameters, to provide a basis for subsequent statistical analysis.

[0060] The statistical analysis module performs statistical analysis on indicators such as cargo throughput, equipment utilization rate, operational efficiency, and operating costs, presenting the results in the form of charts and reports, and predicting the future operation of the terminal based on historical data. The implementation method of the statistical analysis module is as follows:

[0061] Data such as cargo throughput, equipment utilization, operational efficiency, and operating costs are extracted from the database. For cargo throughput, the number of goods entering and leaving is statistically analyzed by time period (day, week, month, etc.). Equipment utilization is calculated by the ratio of actual working time to planned working time. Operational efficiency is calculated based on the number of tasks completed per unit time. Operating cost statistics include equipment maintenance costs, personnel wages, energy consumption, and other expenses. Data visualization technologies, such as the ECharts library, are used to display the statistical results in the form of charts (bar charts, line charts, pie charts) and reports, making it convenient for managers to intuitively understand the station's operational status.

[0062] Time series analysis algorithms (such as the ARIMA model) and machine learning algorithms (such as the Long Short-Term Memory Network LSTM) are used to train historical data to establish a terminal operation prediction model. Based on the trends and patterns of historical data, indicators such as cargo throughput, equipment demand, and workload in the future are predicted to provide a reference for terminal resource allocation and operation decisions. For example, equipment maintenance plans and personnel configuration can be arranged in advance based on the prediction results.

[0063] The risk prediction module uses machine learning algorithms to build a risk prediction model, predicting risks such as cargo damage, equipment failure, and transportation delays, generating risk warnings, and sending them to management personnel and the decision support module. The implementation method of the risk prediction module is as follows:

[0064] Collect a large amount of historical data related to cargo damage, equipment failure, and transportation delays, including cargo attributes, equipment operating parameters, and transportation environment information; use machine learning algorithms (such as random forest algorithm and support vector machine algorithm) to train the data and build cargo damage risk prediction model, equipment failure risk prediction model, and transportation delay risk prediction model respectively; during the training process, optimize the model parameters through methods such as cross-validation to improve the prediction accuracy of the model;

[0065] The system acquires data transmitted from IoT sensing modules and other modules in real time, inputs the data into the corresponding risk prediction model for calculation, and generates detailed risk warning information when the model predicts a high risk (such as the risk of damage to goods under current storage conditions, impending equipment failure, or potential delays in transportation). This includes the risk type, risk level, and possible occurrence time. The warning information is then sent to the terminal devices (such as mobile phones and computers) and decision support modules of relevant management personnel so that timely countermeasures can be taken.

[0066] The decision support module receives risk warning information, decision suggestions, and statistical analysis results. It then provides decision-making solutions based on the actual operation and management objectives of the site, and supports decision simulation and evaluation. The implementation method of the decision support module is as follows:

[0067] The system receives risk warnings from the risk prediction module, decision suggestions from the data processing module, and statistical analysis results from the statistical analysis module. It then combines these with the actual operational status of the site (such as current tasks and resource allocation) and management objectives (such as improving efficiency, reducing costs, and ensuring safety). Using multi-criteria decision analysis methods (such as the Analytic Hierarchy Process, AHP), it comprehensively evaluates various possible decision-making schemes and generates scientifically sound and reasonable solutions. For example, in response to equipment failure risk warnings, the decision support module can generate decision-making schemes such as emergency equipment repair and temporary allocation of replacement equipment.

[0068] Simulation techniques (such as discrete event simulation) are used to simulate the generated decision-making schemes, simulating the station's operational situation after the implementation of the schemes, such as changes in work processes, resource utilization, and cost-effectiveness. By comparing the simulation results of different decision-making schemes, the schemes are evaluated from multiple dimensions (such as efficiency improvement, cost reduction, and risk control), providing management with a basis for decision-making. Management can adjust and optimize the decision-making schemes based on the evaluation results, and finally send the determined decision instructions to the relevant execution modules for execution.

[0069] The communication module adopts wireless network communication protocol and Bluetooth communication protocol to realize stable data transmission and information sharing between modules as well as with remote control terminal and customs supervision system.

[0070] The user interaction module supports managers in querying data, issuing operation commands, and setting system parameters. It displays the real-time operating status of the site, statistical analysis results, and risk warning information, and receives feedback from managers to optimize system functions and user experience.

[0071] The system management module manages system user permissions, data security, and system logs.

[0072] Example 2

[0073] Please see Figure 1 This is the second embodiment of the present invention, which is based on the previous embodiment, but differs in that:

[0074] The user interaction module is implemented as follows:

[0075] The user interface is developed using web front-end technologies (such as HTML5, CSS3, and JavaScript), providing managers with access to functions such as data query, operation command issuance, and system settings. For data query, managers can quickly obtain relevant data by inputting query criteria (such as time range, container number, cargo type, etc.), which is then displayed in tables, charts, and other formats. For operation command issuance, managers can select specific tasks (such as dispatching vehicles, arranging container operations, etc.) on the interface, fill in relevant parameters, and submit the command; the system will then transmit the command to the corresponding module for execution. Regarding system settings, managers can configure user permissions and system parameters.

[0076] The system displays the real-time operational status of the terminal (such as container presence, vehicle location, and equipment operating status), statistical analysis results, and risk warning information in an intuitive way on the user interface. For example, it displays the real-time location of vehicles and containers using a GIS map, and uses different colors and icons to represent equipment operating status. Statistical analysis results are displayed using dynamic charts. Risk warning information is displayed through pop-ups, flashing prompts, and other methods. At the same time, it receives feedback from management personnel, such as suggestions on system functions and problems encountered during operation. The feedback information is recorded and transmitted to the system management module for analysis and processing, and system functions and user experience are optimized based on the feedback information.

[0077] Although embodiments of the invention have been shown and described in detail above, 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 site management system based on the Internet of Things, characterized in that: include The Internet of Things (IoT) sensing module includes RFID tags, temperature and humidity sensors, weight sensors, position sensors, and cameras. The RFID tags identify containers, the temperature and humidity sensors, weight sensors, and position sensors collect cargo information, and the cameras collect image information within the terminal. The data processing module cleans, classifies, and stores the collected information for preprocessing. It uses big data analytics and machine learning algorithms to analyze and mine information on cargo flow patterns, equipment failure modes, and transportation efficiency. The on-site management module obtains container arrival information through the IoT sensing module to update the on-site status, verifies container departure information to update system records, and tracks vehicle entry, exit, parking and cargo loading information in real time to optimize scheduling. The warehouse management module plans storage areas based on the type, attributes, and storage requirements of goods, monitors the storage status of goods through the Internet of Things sensing module, and optimizes the sorting path of goods based on order information and data analysis results. The container operation management module uses optimization algorithms to generate operation scheduling plans based on the operation tasks and the actual situation of the terminal, arranges operation equipment and personnel to perform container unloading, container lifting and container moving operations, and monitors the operation process in real time; The statistical analysis module performs statistical analysis on indicators such as cargo throughput, equipment utilization, operational efficiency, and operating costs, presents the results in the form of charts and reports, and predicts the future operation of the station based on historical data. The risk prediction module uses machine learning algorithms to build a risk prediction model, predict the risks of cargo damage, equipment failure, and transportation delays, generate risk warning information, and send it to management personnel and the decision support module. The decision support module receives risk warning information, decision suggestions, and statistical analysis results, and provides decision-making solutions based on the actual operation and management objectives of the site, supporting decision simulation and evaluation. The communication module adopts wireless network communication protocol and Bluetooth communication protocol to realize stable data transmission and information sharing between modules and between remote control terminal and customs supervision system; The user interaction module supports managers in querying data, issuing operation commands, and setting system parameters. It also displays the real-time operating status of the site, statistical analysis results, and risk warning information, and receives feedback from managers to optimize system functions and user experience. The system management module manages system user permissions, data security, and system logs.

2. The site management system based on the Internet of Things according to claim 1, characterized in that: The data processing module is implemented as follows: After receiving the raw data transmitted by the IoT sensing module, the system performs data cleaning; by setting data format rules, it removes or corrects data that does not meet the standards; it uses data filling algorithms to supplement missing data; and it stores the data in a distributed database according to data type. Big data analytics is used to perform in-depth processing on categorized and stored data; the MapReduce framework is used for parallel computing of massive amounts of data; clustering algorithms are used to analyze the driving trajectory and loading status of transportation vehicles to calculate the utilization efficiency of transportation vehicles; and the valuable information obtained from the analysis and mining is organized into decision suggestions and transmitted to the decision support module in the form of structured data.

3. The site management system based on the Internet of Things according to claim 2, characterized in that: The data filling algorithms include mean filling and linear interpolation.

4. The site management system based on the Internet of Things according to claim 1, characterized in that: The on-site management module is implemented as follows: When a container enters the terminal area, the RFID reader installed at the terminal entrance reads the RFID tag information on the container, obtains the container number, cargo type, and arrival time information, and transmits this information to the on-site management module. The on-site management module updates the on-site status of the container in the system to "arrived" based on the received information, and records the relevant information to the database. When the container leaves the site, the RFID tag is read again at the exit to check whether the exit information is consistent with the system record. If they are consistent, the on-site status of the container is updated to "left the site", and the exit time information is recorded. License plate recognition cameras are installed at the station entrances and exits and in the corresponding areas to collect vehicle entry and exit information in real time, including license plate number and entry and exit time; By connecting with the transportation management system, the system obtains information on the cargo loaded on vehicles. The on-site management module uses geographic information system technology to track and display the driving trajectory and parking location of vehicles in the station in real time. Based on historical vehicle scheduling data and the current station operation status, a genetic algorithm is used to calculate the optimal vehicle scheduling plan.

5. A site management system based on the Internet of Things according to claim 1, characterized in that: The warehouse management module is implemented as follows: Before goods are put into storage, the warehouse management module uses a space planning algorithm to generate the optimal storage area allocation scheme based on the type, attributes and storage requirements of the goods and the spatial layout of the warehouse. The warehouse is divided into different storage areas and each area is assigned a unique identifier. The temperature and humidity sensors and location sensors in the IoT sensing module collect real-time storage status information of goods and transmit the data to the warehouse management module. The warehouse management module analyzes the data and issues an alert if any abnormality is detected. In the goods sorting process, the list of goods to be sorted is determined based on the order information. Combining the goods storage location and data analysis results, the path planning algorithm is used to generate the shortest sorting path to guide the sorting personnel to complete the sorting operation.

6. A site management system based on the Internet of Things according to claim 5, characterized in that: The abnormal situations include excessively high temperature and cargo displacement exceeding the set range.

7. A site management system based on the Internet of Things according to claim 1, characterized in that: The container operation management module is implemented as follows: When a new container operation task is received, the container operation management module first obtains the current situation of the terminal, including the usage status of the operation equipment, the work arrangement of the operation personnel, and the location information of the containers on site. Then, using a mixed integer programming algorithm, with the goal of minimizing job time and maximizing equipment utilization, an optimal job scheduling scheme is generated to determine the job equipment, job sequence, and job time parameters. During the operation, sensors and cameras installed on the equipment collect real-time data on the equipment's operating status and the container's location, which is then transmitted to the container operation management module. The module monitors the operation in real time, and if any abnormality is detected, it immediately issues an alarm, suspends the relevant operation, notifies maintenance or management personnel for handling, and records all data during the operation.

8. A site management system based on the Internet of Things according to claim 1, characterized in that: The statistical analysis module is implemented as follows: Extract statistical data on cargo throughput, equipment utilization, operational efficiency, and operating costs from the database; and use data visualization technology to present the statistical results in the form of charts and reports. Time series analysis and machine learning algorithms are used to train historical data to establish a terminal operation prediction model; based on the trends and patterns of historical data, cargo throughput, equipment demand, and workload indicators are predicted.

9. A site management system based on the Internet of Things according to claim 8, characterized in that: The cargo throughput is calculated by counting the number of goods entering and leaving the site over a period of time; equipment utilization rate is calculated by the ratio of actual working time to planned working time; operational efficiency is calculated based on the number of tasks completed per unit of time; and operating cost statistics include equipment maintenance costs, personnel wages, and various energy consumption expenses.

10. A site management system based on the Internet of Things according to claim 1, characterized in that: The risk prediction module is implemented as follows: Historical data related to cargo damage, equipment failure, and transportation delays were collected, and machine learning algorithms were used to train the data to construct cargo damage risk prediction models, equipment failure risk prediction models, and transportation delay risk prediction models, respectively. During the training process, the model parameters were optimized using cross-validation. The system acquires data transmitted from IoT sensing modules and other modules in real time, inputs the data into the corresponding risk prediction model for calculation, and generates detailed risk warning information when the model predicts a high risk. The warning information is then sent to the terminal devices and decision support modules of management personnel.

Images