A port tallying early warning method, system, device and storage medium

Abstract

The application discloses a port tally early warning method, system, device and storage medium, the method comprises the following steps: a target port tally data center is constructed, and customs declaration data, cargo owner bill of lading data, cargo associated document data, port equipment data, port operation personnel data and port weather data associated with the target port are dynamically integrated and updated; in combination with the real-time data of the target port tally data center, basic information monitoring is carried out, and basic information early warning is triggered according to the basic information monitoring result; in combination with the real-time data of the target port tally data center and the basic information monitoring result, operation efficiency monitoring is carried out, and operation efficiency early warning is triggered according to the operation efficiency monitoring result; in combination with the real-time data of the target port tally data center and the basic information monitoring result, quantity and weight monitoring is carried out, and quantity and weight early warning is triggered according to the quantity and weight monitoring result. The application can realize comprehensive and accurate monitoring and early warning of port tally.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of port tally data monitoring and early warning, in particular to a port tally early warning method, system, device and storage medium. BACKGROUND

[0002] At present, the field of port logistics management is developing rapidly, and port tally operation as a key link has an important influence on the overall operation efficiency and safety of the port. In the cargo tally operation of various ports such as Yangtze River ports and coastal ports, and in the cargo counting and checking link in international freight forwarding services and the transfer point tally management in multimodal transport, the accuracy and timeliness of the tally work are crucial.

[0003] Current port tally operation relies on manual checking and experience-based judgment, which not only leads to frequent information errors, but also causes a series of problems such as low efficiency, poor quantity accuracy, frequent quality problems, improper allocation of equipment resources, and safety compliance risks. In order to further improve the timeliness and efficiency of tally early warning, the existing port tally early warning system also selects to combine intelligent monitoring equipment for monitoring and early warning, including monitoring the working state of loading and unloading equipment through sensors to understand the operation of the equipment and determine whether there are problems affecting the efficiency of the tally; and tracking the location of the goods using RFID technology to master the flow dynamics of the goods. These technologies mainly focus on operation efficiency monitoring, and the monitoring accuracy is also limited. In addition, some port tally early warning systems have made preliminary attempts based on the Internet of Things and big data, but they have not been able to fully cover the monitoring and early warning needs of the tally operation.

[0004] Therefore, the current tally early warning technology lacks comprehensive monitoring of multiple dimensions such as basic information, operation efficiency, quantity and weight, and the monitoring accuracy of each dimension is limited, which leads to the existing port tally early warning being not timely, the problem being discovered late, and thus affecting the overall operation efficiency and safety of the port, and being unable to effectively meet the growing needs of port tally operation. SUMMARY

[0005] In order to realize comprehensive and accurate monitoring of port tally, the present application provides a port tally early warning method, system, device and storage medium.

[0006] In a first aspect, the present application provides a port tally early warning method, comprising:

[0007] constructing a target port tally data platform, dynamically integrating and updating customs declaration data, cargo owner bill of lading data, cargo associated document data, port equipment data, port operation personnel data and port weather data associated with the target port;

[0008] The real-time data of the target port tally data center is combined to perform basic information monitoring, including real-time collection of basic information of the cargo, including identity information, ownership and destination information, document information, cargo size and type, and consistency comparison with the basic information recorded in the customs declaration data, the cargo owner's bill of lading data, and the cargo-related document data in the target port tally data center; when the consistency comparison fails, a corresponding basic information warning is triggered;

[0009] The real-time data of the target port tally data center is combined with the basic information monitoring results to perform operation efficiency monitoring, including real-time monitoring of the time consumption of a single tallying operation, the unprocessed progress of batch tallying operations, and the efficiency of the connection of the tallying operation process, and comparison analysis with the corresponding preset reference value; when the comparison exceeds the corresponding preset reference value, a corresponding operation efficiency warning is triggered; the preset reference value is obtained by querying the time consumption value of a single tallying operation, the unprocessed progress value of batch tallying operations, and the efficiency value of the connection of the tallying operation process of the historical target port under the same port equipment data, port operation personnel data, port weather data, and basic information through consistency comparison of the cargo size and type;

[0010] The real-time data of the target port tally data center is combined with the basic information monitoring results to perform quantity and weight monitoring, real-time acquisition of the actual quantity and weight of the cargo, difference calculation with the quantity and weight of the cargo obtained according to the customs declaration data, the cargo owner's bill of lading data, and the cargo-related document data, and comparison analysis with the corresponding first preset difference threshold; when the comparison exceeds the corresponding first preset difference threshold, a corresponding quantity and weight warning is triggered; the first preset difference threshold is determined by comprehensive calculation of the historical quantity or weight difference threshold influence value and the preset basic difference threshold through query under the same port weather data and the cargo type through consistency comparison in the basic information.

[0011] By adopting the above scheme, the port multi-source data is dynamically integrated to realize comprehensive and real-time monitoring and warning of the basic information of the cargo, the operation efficiency, and the quantity and weight, improve the accuracy and efficiency of the port tallying, and reduce disputes and losses caused by information errors, operation delays, quantity inconsistencies, and the like.

[0012] Preferably, the basic information monitoring includes:

[0013] The current port cargo handling operation scenario is determined based on port equipment data, port personnel data, and port meteorological data, including: normal operation scenario, equipment and personnel downgrade operation scenario, severe weather operation scenario, and a combination of equipment and personnel downgrade and severe weather operation scenario; a preset visual recognition device adapted to the current operation scenario is matched for different operation scenarios, and a preset visual recognition device adapted to each operation scenario is set; the preset visual recognition device can integrate one or more basic information extraction models constructed using deep learning algorithms with different structures and parameters.

[0014] By adopting the above scheme, the current port cargo handling operation scenario can be determined based on port equipment data, port operation personnel data, and port meteorological data. Various scenarios can be accurately identified, and pre-set visual recognition devices can be matched and adapted for different operation scenarios. By utilizing the basic information extraction model integrated therein, basic cargo information can be accurately collected in different scenarios, thereby improving the accuracy and adaptability of basic information monitoring.

[0015] Preferably, the monitoring of operational efficiency includes:

[0016] Predefine contingency events and triggering conditions, and determine the directly and indirectly affected stages of tallying operations for each contingency event; the contingency events include pre-operation contingency events, during-operation contingency events, and post-operation contingency events; pre-operation contingency events include regional yard oversaturation, vessel arrival delays, and temporary changes in policies or regulations; during-operation contingency events include sudden cargo damage and sudden environmental disturbances; post-operation contingency events include cargo delays and yard congestion, and external supply chain disruptions;

[0017] The system monitors in real time whether an emergency event is triggered. Upon detection of an emergency event, it identifies the directly affected and indirectly affected stages of the cargo handling operation. For the directly affected stages, it determines the extended operation time based on the emergency event and generates corresponding stage operation time adjustment durations to adjust the real-time monitoring of the time consumed by a single cargo handling operation and the unprocessed progress of batch cargo handling operations. For the indirectly affected stages, it matches a corresponding emergency delay prediction model based on the identified emergency event type. It determines the extended operation time based on the matched emergency delay prediction model and generates corresponding stage operation time adjustment durations to adjust the real-time monitoring of the time consumed by a single cargo handling operation and the unprocessed progress of batch cargo handling operations. Several emergency delay prediction models exist, each adapting to a type of emergency event. Different emergency delay prediction models employ deep learning algorithms with different structures or parameters and are all trained and generated using historical operation extension durations corresponding to the corresponding types of emergencies.

[0018] By adopting the above scheme, we can predefine emergencies and triggering conditions, determine their direct and indirect impacts on the cargo handling stage, and comprehensively consider all kinds of emergencies that may occur at different stages. We can monitor emergencies in real time and deal with them separately for the direct and indirect impact stages, and flexibly adjust the operation time according to the actual situation.

[0019] Preferably, the quantity and weight monitoring includes:

[0020] Based on the monitoring results of basic information, the scale and type of goods are determined. Quantity and weight monitoring strategies are matched according to the determined scale and type of goods to obtain the actual quantity and weight of goods in real time, as well as whether there are any deviations in actual quantity or weight. When the comparison exceeds the corresponding first preset difference threshold or there is an actual quantity or weight deviation, a corresponding quantity and weight warning is triggered. The quantity and weight monitoring strategies include: single or combined monitoring methods among goods counting and weight monitoring, goods physical condition monitoring, and goods transportation route monitoring. Different goods scales and types are matched with pre-set quantity and weight monitoring strategies that are suitable for them. The goods counting and weight monitoring adopts a comprehensive monitoring method using sensors and visual recognition devices. The goods physical condition monitoring adopts a monitoring method using a visual recognition device that integrates deep learning algorithms to identify whether the goods are damaged or deformed. The goods transportation route monitoring adopts a monitoring method using visual recognition devices to monitor whether there are any missing goods along the goods transportation route.

[0021] By adopting the above scheme, the scale and type of goods are determined based on the monitoring results of basic information, and appropriate quantity and weight monitoring strategies are matched to accurately obtain the actual quantity and weight of goods and detect deviations, triggering early warnings in a timely manner.

[0022] Preferred options also include:

[0023] By combining real-time data, basic information monitoring results, and operational efficiency monitoring results from the target port's tallying data platform, resource allocation monitoring is performed. This includes: acquiring real-time resource allocation plans, which are pre-generated resource allocation schemes based on LSTM neural networks to predict future cargo arrivals, equipment load, personnel attendance rates, and operational efficiency; analyzing whether the real-time resource allocation plans meet the resource allocation requirements under the conditions of cargo arrivals, equipment load, personnel attendance rates, and operational efficiency obtained from the target port's tallying data platform; if the requirements are not met, a resource allocation warning is generated, and a resource allocation plan that meets the current resource allocation requirements is queried from the historical resource allocation plan database based on the acquired cargo arrivals, equipment load, personnel attendance rates, and operational efficiency, and the resource allocation plan is adjusted according to the query results.

[0024] By combining real-time data from the target port's tallying data platform with the basic information monitoring process and operational efficiency monitoring process, safety and compliance monitoring is conducted. This includes: acquiring video stream data collected during the basic information monitoring and operational efficiency monitoring processes; analyzing violations in the video stream data using image analysis technology; generating safety and compliance warnings; recording the safety and compliance recovery process; marking the current basic information monitoring; and after safety and compliance recovery, re-collecting the basic information of the cargo and performing consistency comparisons. Additionally, marking the current operational efficiency monitoring, and eliminating operational delays caused by the safety and compliance recovery process during real-time monitoring of the time consumed in a single tallying operation, the unprocessed progress of batch tallying operations, and the timeliness of the tallying operation process connections.

[0025] By adopting the above scheme, dynamic adjustment of port tallying resources can be achieved, avoiding inefficiency caused by unreasonable resource allocation. At the same time, violations of safety and compliance can be detected and dealt with, reducing the impact of safety and compliance issues on tallying operations and improving the overall efficiency and safety of port tallying operations.

[0026] Preferred options also include:

[0027] Construct a multi-target port cargo handling data exchange platform to determine the relationships between multi-target ports, including: transshipment relationships and distribution relationships;

[0028] During the basic information monitoring of multi-target port tallying related to transshipment, the basic information of goods collected in real time for each port is compared for consistency. If the consistency comparison fails, the corresponding basic information warning is triggered. When monitoring the quantity and weight of multi-target port tallying related to transshipment, it is checked whether the difference between the actual quantity and weight of goods obtained in real time from any two target ports is greater than the corresponding second preset difference threshold. If the difference between the actual quantity and weight of goods obtained in real time from any two target ports is greater than the corresponding second preset difference threshold, the corresponding quantity and weight warning is triggered. When monitoring the operation efficiency of multi-target port tallying related to transshipment, it is checked whether at least two target ports have a similarity of a preset benchmark value greater than a preset similarity. If the same is found, the time consumption of a single tallying operation, the unprocessed progress of batch tallying operations, and the connection time of the tallying operation process are compared between any two target ports. If the difference is greater than the preset difference, the corresponding operation efficiency warning is triggered.

[0029] During the basic information monitoring of multi-target port tallying for cargo with distribution relationships, the system collects cargo identity information, ownership and destination information in real time for each port and performs consistency comparison. It also collects document information in real time and performs consistency comparison after merging cargo size and type according to distribution relationships. If the consistency comparison fails, a corresponding basic information warning is triggered. When monitoring the quantity and weight of cargo tallying at multi-target ports with distribution relationships, the system checks whether the difference between the actual quantity and weight of cargo obtained in real time from target ports with distribution relationships and the difference after merging distribution relationships exceeds a corresponding second preset difference threshold. When the difference between the actual quantity and weight of goods obtained in real time from target ports with distribution relationships and the difference between the distribution and merging results after distribution according to the distribution relationships is greater than the corresponding second preset difference threshold, a corresponding quantity and weight warning is triggered. When monitoring the operational efficiency of cargo handling at multiple target ports with distribution relationships, it is checked whether there are at least two target ports with a preset benchmark similarity greater than the preset similarity. When such a query is found, the time consumption of a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connection at the ports after distribution and merging according to the distribution relationships are compared. If the difference is greater than the preset difference, a corresponding operational efficiency warning is triggered.

[0030] By adopting the above scheme, comprehensive monitoring of tallying operations under different relationships between multiple target ports can be carried out. When anomalies occur in basic information, quantity and weight, operational efficiency, etc., early warnings can be triggered in a timely manner to ensure the accuracy, efficiency and coordination of tallying operations at multiple target ports and reduce operational risks and losses caused by inconsistencies in information, differences in quantity, inefficiency and other problems.

[0031] Preferred options also include:

[0032] A multi-target port cargo handling data exchange platform is constructed to determine the relationships between multiple target ports. The process of establishing the transit time sequence for transshipment relationships and the distribution sequence for distribution relationships is outlined. Following the sequence of transshipment or distribution relationships, during the monitoring of basic information, operational efficiency, and quantity / count data for the preceding target port, if a warning is issued for basic information monitoring, operational efficiency monitoring, or quantity / count data for the preceding target port, the corresponding port's cargo handling data is flagged. This improves the monitoring frequency for each aspect during the monitoring of basic information, operational efficiency, and quantity / count data for the next target port.

[0033] By adopting the above scheme, the order of transshipment and distribution relationships among multiple target ports is sorted out. When a warning is issued at the first target port, the tallying data is marked, thereby increasing the monitoring frequency of the next target port, enhancing the overall monitoring of tallying operations at multiple target ports, timely identifying and handling potential problems, and ensuring the smooth operation of port tallying operations.

[0034] Secondly, this application provides a port cargo handling early warning system, comprising:

[0035] The cargo handling data platform construction module is used to build a cargo handling data platform for the target port, and dynamically integrate and update customs declaration data, cargo owner bill of lading data, cargo-related document data, port equipment data, port operation personnel data, and port weather data associated with the target port.

[0036] The cargo handling basic information early warning module is used to monitor basic information by combining real-time data from the target port's cargo handling data platform. This includes real-time collection of basic cargo information, such as cargo identity information, ownership and destination information, document information, cargo size and type, and comparing it with the basic information recorded in the customs declaration data, cargo owner's bill of lading data, and cargo-related document data in the target port's cargo handling data platform. If the consistency comparison fails, the corresponding basic information early warning is triggered.

[0037] The cargo handling operation efficiency early warning module is used to monitor operation efficiency by combining real-time data and basic information monitoring results from the target port's cargo handling data platform. This includes real-time monitoring of the time consumed by a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling operation process connections, and comparing these with corresponding preset benchmark values. If the comparison shows that the value exceeds the corresponding preset benchmark value, a corresponding operation efficiency early warning is triggered. The preset benchmark values ​​are obtained by querying the historical data of the target port's single cargo handling operation time, unprocessed progress of batch cargo handling operations, and timeliness of cargo handling operation process connections under the condition of consistent comparison of cargo size and type in the same port equipment data, port operation personnel data, port meteorological data, and basic information.

[0038] The cargo handling quantity and weight early warning module is used to monitor the quantity and weight by combining real-time data and basic information monitoring results from the target port's cargo handling data platform. It obtains the actual quantity and weight of the cargo in real time and calculates the difference between the actual quantity and weight and the quantity and weight obtained from customs declaration data, cargo owner's bill of lading data, and cargo-related document data. The difference is then compared with the corresponding first preset difference threshold. If the comparison exceeds the corresponding first preset difference threshold, the corresponding quantity and weight early warning is triggered. The first preset difference threshold is determined by comprehensively calculating the historical quantity or weight difference threshold influence value obtained by querying cargo types that have passed consistent comparison in the same port meteorological data and basic information, and the preset basic difference threshold.

[0039] By adopting the above scheme, a data platform for tallying at the target port is constructed to dynamically integrate and update various types of data, providing comprehensive and real-time data support for subsequent monitoring; basic information monitoring can collect relevant cargo information in real time and compare it with the data in the platform, triggering basic information early warnings to avoid operational errors caused by information errors; operational efficiency monitoring can monitor the time consumption, progress and connection time of tallying operations in real time, and compare it with preset benchmark values, triggering operational efficiency early warnings to improve port tallying efficiency; quantity and weight monitoring can obtain the actual quantity and weight of cargo in real time and compare it with the calculated value, triggering quantity and weight early warnings to ensure the accuracy of quantity and weight.

[0040] Thirdly, this application provides a computer-readable storage medium including a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to perform the method described above.

[0041] Fourthly, this application provides a computer device, the computer device including a memory, a processor and a program stored in the memory and executable thereon, the program being executed by the processor to implement the steps of the method described above.

[0042] In summary, this application has the following beneficial effects:

[0043] 1. By constructing a target port tallying data platform for basic information monitoring, real-time comparison of basic cargo information, and timely triggering of early warnings, information errors can be avoided, and real-time monitoring of basic information anomalies can be achieved. Combining the real-time data of the target port tallying data platform with the basic information monitoring results, operational efficiency can be monitored, and operation time and process connection can be accurately monitored and warned, thereby improving port tallying efficiency. Quantity and weight can be monitored, and counting and weight results can be effectively monitored to ensure the accuracy of quantity and weight, thus achieving comprehensive and accurate monitoring of port tallying.

[0044] 2. Resource allocation monitoring is conducted by combining real-time data from the target port's tallying data platform, basic information monitoring results, and operational efficiency monitoring results. Based on LSTM neural networks, various monitoring information is predicted and resource allocation schemes are generated. The system analyzes whether the resource allocation plans meet actual needs, generates early warnings when they do not, and adjusts the plans accordingly to achieve reasonable resource allocation. Safety and compliance monitoring is also conducted using the aforementioned monitoring data. Image analysis technology is used to analyze violations in video stream data, generating safety and compliance early warnings, recording the recovery process, re-collecting basic information, and eliminating operational delays caused by the safety and compliance recovery process to ensure the safety and compliance of port tallying operations.

[0045] 3. Construct a multi-target port tallying data exchange platform and determine the transshipment and distribution relationships between multi-target ports to achieve unified management of multi-port tallying operations; combine basic information monitoring, operational efficiency monitoring, and quantity and weight monitoring of multi-target ports for comparison to ensure the accuracy of monitoring results, thereby improving the efficiency of multi-port tallying operations. Attached Figure Description

[0046] Figure 1 This is a flowchart of the port cargo handling early warning method described in a specific embodiment;

[0047] Figure 2 This is a schematic diagram of the port cargo handling early warning system described in a specific embodiment. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] This application mainly adopts the construction of a data middle platform to realize multi-dimensional monitoring and early warning of port tallying, so as to achieve comprehensive monitoring of multi-dimensional data of the tallying operation process, timely early warning of problems, and improve the efficiency and safety of port operation. The following is a further detailed description of this application.

[0050] like Figure 1 As shown in the embodiment of this application, a port cargo handling early warning method is disclosed, including: constructing a target port data platform to monitor basic information, operational efficiency, quantity and weight, etc., so as to realize monitoring and early warning of data based on the data platform from different dimensions.

[0051] S1. Build a cargo handling data platform for the target port and dynamically integrate and update data associated with the target port.

[0052] Specifically, building a data platform for cargo handling at the target port includes data acquisition equipment and data processing servers.

[0053] The data acquisition equipment can utilize various sensors, data interfaces, and visual recognition devices to collect in real-time customs declaration data, cargo owner bill of lading data, cargo-related document data, port equipment data, port personnel data, and port meteorological data associated with the target port. Customs declaration data includes consignor / consignee information (name, address, etc.), cargo information (commodity name and type, specifications, quantity, etc.), trade information (trade method, transaction method, etc.), and transportation information (transportation method, bill of lading / waybill number, etc.). Cargo owner bill of lading data includes bill of lading header information (carrier information, shipper information, etc.), cargo and transportation information (cargo information, transportation information), bill of lading terms, and remarks. Cargo-related document data includes basic cargo information recorded in documents other than customs declarations and bills of lading, such as supplementary cargo information forms. Port equipment data includes port equipment status, quantity, and operational data. Port personnel data includes port personnel attendance rate and operational status data. Specifically, customs declaration data and cargo owner bill of lading data are collected through data interfaces with the customs system and cargo owner information system; port equipment data is collected using sensors to monitor the operating status and location of the equipment; port personnel data is collected through employee attendance systems, positioning devices, etc.; and port meteorological data is obtained from the meteorological department's interface.

[0054] The data processing server is used to dynamically integrate and update the collected data, including cleaning, organizing, and storing the data to form a data platform for cargo handling at the target port. Through the construction of this data platform, centralized management and dynamic updates of various types of data can be achieved, providing accurate data support for subsequent monitoring and early warning.

[0055] S2. Monitor basic information by combining real-time data from the target port's cargo handling data platform.

[0056] Specifically, in order to achieve comprehensive cargo handling data monitoring, basic information of goods arriving at the target port is collected in real time, including: cargo identity information, ownership and destination information, document information, cargo size and type, such as: consignor / consignee, destination port / yard, bill of lading, packing list, customs declaration, container number / document number, cargo type, specifications, etc. This data can all be obtained using visual recognition devices, such as: obtaining identity information by taking pictures of cargo labels and markings with a camera, and obtaining document information by scanning documents with a scanner.

[0057] The real-time collected basic information is compared with the basic information recorded in the target port tallying data, Taichung Customs declaration data, cargo owner bill of lading data, and cargo-related document data. If the consistency comparison fails, such as inconsistencies between document information and physical markings (e.g., altered container number, incorrect cargo type), incorrect entry of destination port, missing key documents, or unclear information (e.g., quantity not marked), the corresponding basic information warning is triggered.

[0058] S3. Combine real-time data from the target port's cargo handling data platform with basic information monitoring results to monitor operational efficiency.

[0059] Specifically, operational efficiency monitoring mainly focuses on real-time monitoring of the time consumed by a single tallying operation, the unprocessed progress of batch tallying operations, and the efficiency of the tallying operation process. Among these, the time consumed by a single tallying operation refers to the duration recorded by a timer from the arrival of goods to the completion of tallying. The unprocessed progress of batch tallying operations refers to the unfinished progress of batch processing (several single tallying operations within a batch) recorded by a counter. The efficiency of the tallying operation process refers to the ratio of the interval time between each stage of the tallying operation to the preset interval time between each stage.

[0060] The time taken for a single tallying operation, the progress of unprocessed batch tallying operations, and the efficiency of the tallying operation process are compared and analyzed with their corresponding preset benchmark values, based on the monitoring data. Considering the significant differences in the time taken for a single tallying operation, the progress of unprocessed batch tallying operations, and the efficiency of the tallying operation process under different tallying operation conditions, to make the comparison more accurate, historical tallying operation data of the target port is incorporated. The preset benchmark values ​​are obtained by querying cargo size and type that have passed consistency comparison in the same port equipment data, port personnel data, port meteorological data, and basic information, and comparing them with the historical target port's time taken for a single tallying operation and the progress of unprocessed batch tallying operations. The system obtains the timeliness values ​​for the connection between cargo handling operations and the cargo handling process. If no identical conditions exist, the system can query the cargo size and type through consistency comparison in port equipment data, port personnel data, port meteorological data, and basic information. If the similarity is greater than the preset similarity (95%), the system can obtain the time consumption value of a single cargo handling operation, the unprocessed progress value of a batch cargo handling operation, and the timeliness value for the connection between cargo handling operations for the target port under the corresponding conditions. If there are multiple time consumption values ​​of a single cargo handling operation, the unprocessed progress value of a batch cargo handling operation, and the timeliness value for the connection between cargo handling operations for the target port, the system can select the value with the highest frequency as the preset benchmark value, such as the unprocessed progress value of a batch cargo handling operation (20%).

[0061] When the comparison exceeds the corresponding preset benchmark value, the corresponding operation efficiency warning is triggered; for example, if the unprocessed progress of the batch sorting operation is greater than 20%, it is determined that the current operation efficiency is low and an unprocessed progress warning for the batch sorting operation is generated.

[0062] S4. Combine the real-time data and basic information monitoring results from the target port's cargo handling data platform to monitor quantity and weight.

[0063] Specifically, quantity and weight monitoring mainly involves obtaining the actual quantity and weight of goods in real time. The actual quantity and weight of goods can be obtained by weighing sensors, counters, or visual recognition devices. For example, weighing sensors can be used to measure the weight of goods, and counters can be used to count the quantity of goods; or visual recognition devices can be used to identify the quantity of goods, thereby calculating the weight of goods.

[0064] The actual quantity and weight of the goods are obtained and then compared with the quantity and weight obtained from customs declaration data, shipper's bill of lading data, and related cargo documents. The difference is then compared with a corresponding first preset difference threshold. The quantity and weight obtained from customs declaration data, shipper's bill of lading data, and related cargo documents include both directly obtaining the quantity and weight recorded in these data and calculating the quantity and weight based on the basic cargo data in these data.

[0065] The difference between the actual quantity of goods and the quantity calculated based on customs declaration data, shipper's bill of lading data, and related cargo documents is compared with a first preset threshold for the quantity difference. If the difference exceeds this threshold, a corresponding quantity warning is triggered. Similarly, the difference between the actual weight of goods and the weight calculated based on customs declaration data, shipper's bill of lading data, and related cargo documents is compared with a first preset threshold for the weight difference. If the difference exceeds this threshold, a corresponding weight warning is triggered. This process takes into account different cargo handling conditions. Under certain circumstances, the weight and quantity of goods may change due to external influencing factors. For example, the weight of goods may increase due to rain during heavy rain, and the quantity recognized by the image may be inaccurate. To monitor the quantity and weight more accurately, a first preset difference threshold is designed. This threshold is determined by querying the historical quantity or weight difference threshold value obtained under the condition of consistent comparison of the same cargo type in port equipment data, port operation personnel data, port meteorological data, and basic information, and comprehensively calculating it with the preset basic difference threshold. For example, under the first specific condition, the historical quantity difference threshold value is 3% of the preset basic difference threshold X, and the first preset difference threshold corresponding to the calculated quantity is 3%*X.

[0066] In addition to monitoring basic information, operational efficiency, and quantity and weight, the method also includes, in order to achieve more comprehensive monitoring of port tallying operations:

[0067] S5. Combine the real-time data, basic information monitoring results, and operational efficiency monitoring results from the target port's cargo handling data platform to monitor resource allocation.

[0068] Specifically, resource allocation monitoring mainly focuses on whether the allocated cargo arrival volume A, the allocated operating equipment B, and the operating personnel C can meet the pre-set cargo handling efficiency D (e.g., 80%). To achieve more accurate resource allocation, a resource allocation plan can be pre-designed based on predicted future cargo arrival volumes, equipment load, personnel attendance rate, and operating efficiency. This serves as a guide for resource allocation. Therefore, resource allocation monitoring prioritizes obtaining real-time resource allocation plans to determine whether they meet actual resource allocation requirements. These real-time resource allocation plans are pre-generated based on LSTM neural networks that predict future cargo arrival volumes, equipment load, personnel attendance rate, and operating efficiency. They include: equipment operation plans, personnel attendance requirements, and operating efficiency requirements.

[0069] Analyze whether the real-time resource allocation plan meets the resource allocation requirements based on the real-time data, basic information monitoring results, and operational efficiency monitoring results obtained from the target port's cargo handling data platform, including cargo arrival volume, equipment load, personnel attendance rate, and operational efficiency. In other words, whether the obtained cargo arrival volume meets the operational efficiency requirements under the equipment operation plan and personnel attendance requirements in the allocation plan.

[0070] If the conditions are not met, a resource allocation warning is generated, and a resource allocation plan that meets the current resource allocation requirements is queried from the historical resource allocation plan database based on the obtained cargo arrival volume, equipment load, personnel attendance rate and operational efficiency. The resource allocation plan is then adjusted according to the query results. The historical resource allocation plan database stores the equipment load and personnel attendance rate corresponding to the current arrival volume for the target port in order to meet the operational efficiency requirements.

[0071] S6. Combine the real-time data, basic information monitoring process, and operational efficiency monitoring process of the target port's cargo handling data platform to conduct safety and compliance monitoring.

[0072] Specifically, safety and compliance monitoring includes: acquiring video stream data collected during basic information monitoring and operational efficiency monitoring; analyzing the video stream data for violations using image analysis technology, such as employees modifying document information during basic information monitoring, or employees not wearing safety helmets or protective equipment during operational efficiency monitoring; generating safety and compliance warnings; recording the safety and compliance recovery process, including the time spent on safety recovery; marking the current basic information monitoring; re-collecting basic cargo information and performing consistency comparison after safety and compliance recovery; and marking the current operational efficiency monitoring, eliminating operational delays caused by the safety and compliance recovery process (i.e., the time spent on safety recovery) during real-time monitoring of the time spent on a single cargo handling operation, the progress of unprocessed batch cargo handling operations, and the timeliness of cargo handling process connections.

[0073] Similarly, combining quantity and weight monitoring with safety and compliance monitoring includes: acquiring video stream data collected during the quantity and weight monitoring process; analyzing the video stream data for violations, such as employees maliciously modifying data, using image analysis technology; generating safety and compliance warnings; recording the safety and compliance recovery process, including the time spent on safety recovery; marking the current basic information monitoring accordingly; and re-collecting the quantity and weight of goods in real time after safety and compliance recovery to complete accurate quantity and weight monitoring.

[0074] By adopting the above methods, it is possible to monitor and issue early warnings for port tallying operations from multiple dimensions, including basic information, operational efficiency, quantity and weight, resource allocation, and security compliance, based on data from a data platform. When an anomaly occurs in any dimension of the data, a corresponding early warning is triggered in a timely manner, reminding relevant personnel to take measures, thus achieving comprehensive multi-dimensional monitoring of port tallying operations.

[0075] In a specific embodiment, considering that different port tallying operation scenarios can affect the collection and recognition of basic information, by matching appropriate visual recognition devices, the method can better adapt to different operating environments and improve the accuracy and reliability of basic information monitoring; the basic information monitoring in the method includes:

[0076] The current port tallying operation scenario is determined based on port equipment data, port personnel data, and port meteorological data, including: normal operation scenario, equipment and personnel downgrade operation scenario, severe weather operation scenario, and a combination of equipment and personnel downgrade and severe weather operation scenario; among them, a port tallying operation scenario recognition model can be constructed using a neural network algorithm, which is trained and generated using historical port equipment data, port personnel data, and port meteorological data labeled with the operation scenario.

[0077] Pre-set visual recognition devices are matched to different work scenarios and adapted to the current work scenario, with each work scenario having its own pre-set visual recognition device. These pre-set visual recognition devices can integrate one or more basic information extraction models constructed using deep learning algorithms with different structures and parameters. The deep learning algorithms are neural networks with different structures or parameters, such as different numbers of convolutional layers and different convolutional kernel sizes. For normal scenarios, a first pre-set visual recognition device is matched, which integrates a basic information extraction model with a first number of convolutional layers and a first convolutional kernel size.

[0078] In a specific embodiment, during port operations, unforeseen events often occur that reduce operational efficiency. Considering the extended operation time caused by these unforeseen events, and to further improve the accuracy of operational efficiency monitoring, the operational efficiency monitoring in the system includes:

[0079] Predefine contingency events and triggering conditions, and determine the directly affected tallying operation stage (the stage where the contingency event is located or the next stage) and indirectly affected tallying operation stages (adjacent to the stage where the contingency event is located or the next stage) for each contingency event. Contingency events include pre-operation contingency events, in-operation contingency events, and post-operation contingency events. Pre-operation contingency events include regional yard oversaturation, ship arrival delays, and temporary changes in policies or regulations, with corresponding triggering conditions including: regional yard saturation exceeding a preset saturation level, ship arrival time exceeding a preset time, and receiving temporary modifications to customs clearance policies or regulations. In-operation contingency events include sudden cargo damage and sudden environmental disturbances, with corresponding triggering conditions including: cargo breakage and deformation, and sudden changes in meteorological data. Post-operation contingency events include cargo delays and yard congestion, and external supply chain disruptions, with corresponding triggering conditions including: receiving information that cargo needs temporary storage in the yard and the quantity of temporarily stored cargo in the yard exceeds a preset quantity, and receiving information about disruptions in the port's upstream and downstream logistics chains.

[0080] Real-time monitoring to detect and determine whether an emergency is triggered. If an emergency is detected, the directly affected and indirectly affected cargo handling stages are identified. For the directly affected cargo handling stages, the extended operation time can be directly determined based on the delay caused by the emergency, and the corresponding stage operation time adjustment time can be generated to adjust the real-time monitoring of the time consumed by a single cargo handling operation (time consumed by a single cargo handling operation + X) and the unprocessed progress of batch cargo handling operations (1 - (for several single cargo handling operations in the same batch, calculate the sum of the actual time consumed by several single cargo handling operations including the time consumed by a single cargo handling operation + X / the sum of the estimated time consumed by several single cargo handling operations in the same batch)).

[0081] For the indirectly affected cargo handling stage, based on the determined type of emergency event, a corresponding emergency delay prediction model is matched; the extended operation time is determined according to the adapted emergency delay prediction model, and the corresponding stage operation time adjustment time is generated to adjust the real-time monitoring of the time consumption of a single cargo handling operation and the unprocessed progress of batch cargo handling operations; wherein, there are several emergency delay prediction models, each of which is adapted to a type of emergency event, and different emergency delay prediction models adopt deep learning algorithms with different structures or parameters, and all are trained and generated through the corresponding operation extension time of historical emergency events of the corresponding type.

[0082] In a specific embodiment, in addition to conventionally using sensors or visual recognition devices to identify and monitor the quantity and weight of the goods themselves, the method can assist in monitoring the quantity and weight in situations where goods may fall during transportation or where liquid goods packaging is damaged or leaking, thereby further improving the accuracy of quantity and weight monitoring; the quantity and weight monitoring in the method includes:

[0083] Based on the monitoring results of basic information, the scale and type of goods are determined. Quantity and weight monitoring strategies are then matched according to the determined scale and type. These strategies include individual or combined monitoring methods among goods counting and weight monitoring, goods physical condition monitoring, and goods transportation route monitoring. Different goods scales and types are matched with pre-set quantity and weight monitoring strategies. For example, for small-scale goods that are solid, a first quantity and weight monitoring strategy is used, including goods counting and weight monitoring and goods transportation route monitoring. Similarly, for medium-scale goods that are liquid, a second quantity and weight monitoring strategy is used, including goods counting and weight monitoring, goods physical condition monitoring, and goods transportation route monitoring. The goods counting and weight monitoring uses a combination of sensors and visual recognition devices. The goods physical condition monitoring uses a visual recognition device incorporating deep learning algorithms to identify whether the goods are damaged or deformed. The goods transportation route monitoring uses a visual recognition device to monitor whether any goods are missing from the transportation route.

[0084] The system can acquire the actual quantity and weight of goods in real time, as well as whether there are any deviations in the actual quantity or weight. This means that it can detect situations where there is damage or deformation, or where goods are missing from the transportation route. When the comparison exceeds the corresponding first preset difference threshold or there is a deviation in the actual quantity or weight, the system can trigger the corresponding quantity and weight warning.

[0085] Furthermore, during normal cargo transportation, the real-time actual quantity and weight of the cargo are almost identical to those obtained at the previous moment. By comparing the difference between the real-time actual quantity and weight and those obtained at the previous moment, if there is a difference in quantity or weight that exceeds the corresponding third preset difference threshold, it indicates that there is a problem with the current cargo transportation. The current cargo is then marked as a priority cargo, and the matching quantity and weight monitoring strategy is adjusted. The adjusted matching quantity and weight monitoring strategy includes physical status monitoring or cargo transportation path monitoring to achieve more accurate monitoring and increase the monitoring frequency to strengthen the monitoring of cargo quantity and weight in order to promptly detect abnormalities in cargo quantity and weight.

[0086] In a specific embodiment, besides cargo handling alerts for a single target port, in practical applications, multiple target ports are involved in intermodal cargo transport. Interactive verification is achieved through these intermodal transport relationships among multiple target ports to improve the accuracy of multi-dimensional monitoring. The method also includes:

[0087] A multi-target port tallying data exchange platform is constructed to determine the relationships between multiple target ports, including transshipment relationships and distribution relationships. This platform is used to acquire tallying data from other target port tallying data platforms, in addition to its own port tallying data, and then transfer this data back to its own platform. All acquired port tallying data is marked with a number indicating its origin from a target port. Furthermore, the platform can pre-configure the exchange and transmission of tallying data between multiple target ports according to pre-set relationships.

[0088] During the basic information monitoring of multi-target port cargo handling related to transshipment, if the goods carried by multiple target ports are the same batch of goods, then the corresponding basic information is the same, and the corresponding quantity and weight are almost identical. Therefore, a consistency comparison is performed on the basic information of the goods collected in real time for each port. If the consistency comparison fails, a corresponding basic information warning is triggered. When monitoring the quantity and weight of multi-target port cargo handling related to transshipment, it is checked whether the difference between the actual quantity and weight of the goods obtained in real time from any two target ports exceeds the corresponding second preset difference threshold. If the difference between the actual quantity and weight of the goods obtained in real time from any two target ports exceeds the second preset difference threshold, a warning is triggered. When the difference between weights exceeds the corresponding second preset difference threshold, a corresponding quantity and weight warning is triggered. When monitoring the operational efficiency of cargo handling at multiple target ports with transshipment relationships, it is necessary to query whether there are at least two target ports whose corresponding preset benchmark similarity is greater than the preset similarity. If such a query is found, it indicates that there are at least two target ports whose corresponding port equipment data, port personnel data, and port weather data have small differences. The time taken for a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connection are compared between any two target ports. If the difference is greater than the preset difference, a corresponding operational efficiency warning is triggered.

[0089] During the basic information monitoring of multi-target port tallying for cargo with distribution relationships, if it indicates that the cargo carried by the target port before distribution and the cargo consolidated by multiple target ports after distribution are the same batch of cargo, then the corresponding basic information is the same, and the corresponding quantity and weight are almost identical. For each port, the identity information, ownership, and destination information of the cargo are collected in real time for consistency comparison. Document information is also collected in real time, and the cargo size and type are compared after distribution and consolidation according to the distribution relationship (e.g., the size of the cargo recorded before distribution versus the size of the consolidated cargo recorded after distribution). If the consistency comparison fails, a corresponding basic information warning is triggered. When monitoring the quantity and weight of multi-target port tallying for cargo with distribution relationships, it is necessary to query whether there are target ports with distribution relationships that are acquiring cargo in real time. If the difference between the actual quantity and weight after being combined according to the distribution relationship is greater than the corresponding second preset difference threshold, a corresponding quantity and weight warning is triggered when the difference between the actual quantity and weight of the goods after being combined according to the distribution relationship at a target port with a distribution relationship is greater than the corresponding second preset difference threshold. When monitoring the operational efficiency of cargo handling at multiple target ports with distribution relationships, it is checked whether there are at least two target ports with a preset benchmark similarity greater than the preset similarity. When such a query is found, the time consumption of a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connection at the port after being combined according to the distribution relationship are compared. If the difference is greater than the preset difference, a corresponding operational efficiency warning is triggered.

[0090] Furthermore, to further improve monitoring accuracy, during the construction of a multi-target port cargo handling data exchange platform and the determination of the relationships between multiple target ports, the sequential order of transshipment and distribution relationships was analyzed. Based on the sequential order of transshipment or distribution relationships, during the monitoring of basic information, operational efficiency, and quantity / count data for the preceding target port, if a warning is issued for basic information, operational efficiency, or quantity / count data for the preceding target port, the corresponding cargo handling data for that port is flagged. This increases the monitoring frequency for the next target port during the same monitoring processes.

[0091] like Figure 2 As shown in the figure, this application discloses a port cargo handling early warning system, which specifically includes:

[0092] Cargo tallying data platform construction module 101 is used to build a cargo tallying data platform for the target port, dynamically integrating and updating customs declaration data, cargo owner bill of lading data, cargo-related document data, port equipment data, port operation personnel data, and port meteorological data associated with the target port.

[0093] The cargo handling basic information early warning module 102 is used to monitor basic information by combining real-time data from the target port's cargo handling data platform. This includes real-time collection of basic cargo information, such as cargo identity information, ownership and destination information, document information, cargo size and type, and comparing it with the basic information recorded in the customs declaration data, cargo owner's bill of lading data, and cargo-related document data in the target port's cargo handling data platform. If the consistency comparison fails, the corresponding basic information early warning is triggered.

[0094] The cargo handling operation efficiency early warning module 103 is used to monitor operation efficiency by combining real-time data and basic information monitoring results from the target port's cargo handling data platform. This includes real-time monitoring of the time consumed by a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling operation process connections, and comparing and analyzing these against corresponding preset benchmark values. If the comparison shows that the value exceeds the corresponding preset benchmark value, a corresponding operation efficiency early warning is triggered. The preset benchmark values ​​are obtained by querying the historical data of the target port's single cargo handling operation time, unprocessed progress of batch cargo handling operations, and timeliness of cargo handling operation process connections under the condition of consistent comparison of cargo size and type in the same port equipment data, port operation personnel data, port meteorological data, and basic information.

[0095] The cargo handling quantity and weight early warning module 104 is used to combine real-time data and basic information monitoring results from the target port's cargo handling data platform to monitor quantity and weight. It obtains the actual quantity and weight of the cargo in real time and calculates the difference between the actual quantity and weight and the quantity and weight obtained from customs declaration data, cargo owner's bill of lading data, and cargo-related document data. It then compares and analyzes the difference with a corresponding first preset difference threshold. If the comparison exceeds the corresponding first preset difference threshold, a corresponding quantity and weight early warning is triggered. The first preset difference threshold is determined by comprehensively calculating the historical quantity or weight difference threshold influence value obtained by querying cargo types that have passed consistent comparison in the same port meteorological data and basic information, and the preset basic difference threshold.

[0096] In one specific embodiment, the system further includes:

[0097] The resource allocation monitoring and early warning module 105 is used to monitor resource allocation by combining real-time data, basic information monitoring results, and operational efficiency monitoring results from the target port's tallying data platform. This includes: acquiring a real-time resource allocation plan, which is a pre-generated resource allocation scheme based on LSTM neural network predictions of future cargo arrivals, equipment load, personnel attendance rate, and operational efficiency; analyzing whether the real-time resource allocation plan meets the resource allocation requirements under the conditions of cargo arrivals, equipment load, personnel attendance rate, and operational efficiency obtained from the target port's tallying data platform; and generating a resource allocation early warning if the requirements are not met, and querying the historical resource allocation plan database for a plan that meets the current resource allocation requirements based on the acquired cargo arrivals, equipment load, personnel attendance rate, and operational efficiency, and adjusting the resource allocation plan according to the query results.

[0098] The safety and compliance monitoring and early warning module 106 is used to combine real-time data from the target port's tallying data platform with the basic information monitoring process and the operational efficiency monitoring process to conduct safety and compliance monitoring. This includes: acquiring video stream data collected during the basic information monitoring and operational efficiency monitoring processes; analyzing violations in the video stream data using image analysis technology; generating safety and compliance early warnings; recording the safety and compliance recovery process; marking the current basic information monitoring; re-collecting the basic information of the cargo and performing consistency comparison after safety and compliance recovery; and marking the current operational efficiency monitoring. In the process of real-time monitoring of the time consumed by a single tallying operation, the unprocessed progress of batch tallying operations, and the timeliness of the tallying operation process connection, the module eliminates operational delays caused by the safety and compliance recovery process.

[0099] In one specific embodiment, the system further includes:

[0100] The multi-port cargo handling early warning optimization module 107 is used to construct a multi-target port cargo handling data exchange platform, determine the relationships between multi-target ports, including transshipment relationships and distribution relationships; during the basic information monitoring of multi-target port cargo handling for transshipment relationships, a consistency comparison is performed on the basic information of the cargo collected in real time for each port; when the consistency comparison fails, the corresponding basic information early warning is triggered; when monitoring the quantity and weight of multi-target port cargo handling for transshipment relationships, it queries whether the difference between the actual quantity and weight of the cargo obtained in real time from any two target ports is greater than the corresponding second preset difference threshold. When the difference between the actual quantity and weight of goods obtained in real time from any two target ports exceeds the corresponding second preset difference threshold, a corresponding quantity and weight warning is triggered. When monitoring the operational efficiency of cargo handling at multiple target ports with transshipment relationships, it checks whether at least two target ports have a preset benchmark similarity greater than a preset similarity. If found, it compares the time consumed by a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connections between any two target ports. If the difference exceeds the preset difference, a corresponding operational efficiency warning is triggered. For cargo handling at multiple target ports with distribution relationships... During basic information monitoring, for each port, real-time collection of cargo identity information, ownership and destination information is compared for consistency. Real-time collection of documentation information and cargo size and type are also compared after being combined according to the distribution relationship. If the consistency comparison fails, a corresponding basic information warning is triggered. When monitoring the quantity and weight of cargo at multiple target ports with distribution relationships, it checks whether the difference between the actual quantity and weight of cargo obtained in real-time from target ports with distribution relationships and after being combined according to the distribution relationship is greater than the corresponding second preset difference threshold. If a distribution relationship is found... When the difference between the actual quantity and weight of goods obtained in real time from the target port of the relationship and the value after being merged according to the distribution relationship is greater than the corresponding second preset difference threshold, the corresponding quantity and weight warning is triggered; when monitoring the operation efficiency of cargo handling at multiple target ports with distribution relationship, it is checked whether there are at least two target ports with a preset benchmark similarity greater than the preset similarity. When the query is found, the time consumption of a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connection are compared after the distribution relationship is merged. If the difference is greater than the preset difference, the corresponding operation efficiency warning is triggered;

[0101] The multi-port tallying early warning optimization module 107 is also used to construct a multi-target port tallying data exchange platform, determine the process of correlation between multiple target ports, sort out the order of transshipment time in transshipment correlation and the order of distribution in distribution correlation; according to the order of transshipment correlation or distribution correlation, in the process of monitoring basic information, operational efficiency and quantity and count for the previous target port, once there is a basic information monitoring early warning, operational efficiency early warning and quantity and count early warning for the previous target port, the corresponding port tallying data is marked, and the monitoring frequency is increased in the process of monitoring basic information, operational efficiency and quantity and count for the next target port.

[0102] This application also discloses a computer-readable storage medium.

[0103] Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and executed, such as the port cargo handling early warning method described above. The computer-readable storage medium includes, for example, various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0104] This application also discloses a computer device.

[0105] Specifically, the computer device includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and executed to implement the aforementioned port cargo handling early warning method.

[0106] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A port cargo handling early warning method, characterized in that, include: Build a cargo handling data platform for the target port to dynamically integrate and update customs declaration data, cargo owner bill of lading data, cargo-related document data, port equipment data, port operation personnel data, and port meteorological data associated with the target port. By combining real-time data from the target port's cargo handling data platform, basic information monitoring is performed, including real-time collection of basic cargo information, such as cargo identity information, ownership and destination information, document information, cargo size and type, and consistency comparison with the basic information recorded in the customs declaration data, cargo owner's bill of lading data, and cargo-related document data in the target port's cargo handling data platform; if the consistency comparison fails, the corresponding basic information warning is triggered. By combining real-time data from the target port's tallying data platform with basic information monitoring results, operational efficiency is monitored. This includes real-time monitoring of the time consumed by a single tallying operation, the unprocessed progress of batch tallying operations, and the timeliness of the tallying operation process connection, and comparing these with corresponding preset benchmark values. If the comparison exceeds the corresponding preset benchmark value, a corresponding operational efficiency warning is triggered. The preset benchmark values ​​are obtained by querying historical data on the target port's single tallying operation time, unprocessed progress of batch tallying operations, and timeliness of the tallying operation process connection under the condition of consistent comparison of cargo size and type in the same port equipment data, port operation personnel data, port meteorological data, and basic information. Combining real-time data from the target port's cargo handling data platform with basic information monitoring results, quantity and weight monitoring is performed. The actual quantity and weight of the cargo are obtained in real time, and the difference is calculated with the quantity and weight of the cargo obtained from customs declaration data, cargo owner's bill of lading data, and cargo-related document data. The difference is then compared with the corresponding first preset difference threshold. If the comparison exceeds the corresponding first preset difference threshold, the corresponding quantity and weight warning is triggered. The first preset difference threshold is determined by comprehensively calculating the historical quantity or weight difference threshold impact value obtained by querying cargo types that have passed consistent comparison in the same port meteorological data and basic information, and the preset basic difference threshold. The operation efficiency monitoring includes: Predefine contingency events and triggering conditions, and determine the directly and indirectly affected stages of tallying operations for each contingency event; the contingency events include pre-operation contingency events, during-operation contingency events, and post-operation contingency events; pre-operation contingency events include regional yard oversaturation, vessel arrival delays, and temporary changes in policies or regulations; during-operation contingency events include sudden cargo damage and sudden environmental disturbances; post-operation contingency events include cargo delays and yard congestion, and external supply chain disruptions; The system monitors in real time whether an emergency event is triggered. Upon detection of an emergency event, it identifies the directly affected and indirectly affected stages of the cargo handling operation. For the directly affected stages, it determines the extended operation time based on the emergency event and generates corresponding stage operation time adjustment durations to adjust the real-time monitoring of the time consumed by a single cargo handling operation and the unprocessed progress of batch cargo handling operations. For the indirectly affected stages, it matches a corresponding emergency delay prediction model based on the identified emergency event type. It determines the extended operation time based on the matched emergency delay prediction model and generates corresponding stage operation time adjustment durations to adjust the real-time monitoring of the time consumed by a single cargo handling operation and the unprocessed progress of batch cargo handling operations. Several emergency delay prediction models exist, each adapting to a type of emergency event. Different emergency delay prediction models employ deep learning algorithms with different structures or parameters and are all trained and generated using historical operation extension durations corresponding to the corresponding types of emergencies. Construct a multi-target port cargo handling data exchange platform to determine the relationships between multi-target ports, including: transshipment relationships and distribution relationships; During the basic information monitoring of multi-target port tallying related to transshipment, the basic information of goods collected in real time for each port is compared for consistency. If the consistency comparison fails, the corresponding basic information warning is triggered. When monitoring the quantity and weight of multi-target port tallying related to transshipment, it is checked whether the difference between the actual quantity and weight of goods obtained in real time from any two target ports is greater than the corresponding second preset difference threshold. If the difference between the actual quantity and weight of goods obtained in real time from any two target ports is greater than the corresponding second preset difference threshold, the corresponding quantity and weight warning is triggered. When monitoring the operation efficiency of multi-target port tallying related to transshipment, it is checked whether at least two target ports have a similarity of a preset benchmark value greater than a preset similarity. If the same is found, the time consumption of a single tallying operation, the unprocessed progress of batch tallying operations, and the connection time of the tallying operation process are compared between any two target ports. If the difference is greater than the preset difference, the corresponding operation efficiency warning is triggered. During the basic information monitoring of multi-target port tallying for cargo with distribution relationships, the system collects cargo identity information, ownership and destination information in real time for each port and performs consistency comparison. It also collects document information in real time and performs consistency comparison after merging cargo size and type according to distribution relationships. If the consistency comparison fails, a corresponding basic information warning is triggered. When monitoring the quantity and weight of cargo tallying at multi-target ports with distribution relationships, the system checks whether the difference between the actual quantity and weight of cargo obtained in real time from target ports with distribution relationships and the difference after merging distribution relationships exceeds a corresponding second preset difference threshold. When the difference between the actual quantity and weight of goods obtained in real time from target ports with distribution relationships and the difference between the distribution and merging results after distribution according to the distribution relationships is greater than the corresponding second preset difference threshold, a corresponding quantity and weight warning is triggered. When monitoring the operational efficiency of cargo handling at multiple target ports with distribution relationships, it is checked whether there are at least two target ports with a preset benchmark similarity greater than the preset similarity. When such a query is found, the time consumption of a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connection at the ports after distribution and merging according to the distribution relationships are compared. If the difference is greater than the preset difference, a corresponding operational efficiency warning is triggered. A multi-target port cargo handling data exchange platform is constructed to determine the relationships between multiple target ports. The process of establishing the transit time sequence for transshipment relationships and the distribution sequence for distribution relationships is outlined. Following the sequence of transshipment or distribution relationships, during the monitoring of basic information, operational efficiency, and quantity / count data for the preceding target port, if a warning is issued for basic information monitoring, operational efficiency monitoring, or quantity / count data for the preceding target port, the corresponding port's cargo handling data is flagged. This improves the monitoring frequency for each aspect during the monitoring of basic information, operational efficiency, and quantity / count data for the next target port.

2. The port cargo handling early warning method according to claim 1, characterized in that, The basic information monitoring includes: The current port cargo handling operation scenario is determined based on port equipment data, port personnel data, and port meteorological data, including: normal operation scenario, equipment and personnel downgrade operation scenario, severe weather operation scenario, and a combination of equipment and personnel downgrade and severe weather operation scenario; a preset visual recognition device adapted to the current operation scenario is matched for different operation scenarios, and a preset visual recognition device adapted to each operation scenario is set; the preset visual recognition device can integrate one or more basic information extraction models constructed using deep learning algorithms with different structures and parameters.

3. The port cargo handling early warning method according to claim 1, characterized in that, The quantity and weight monitoring includes: Based on the monitoring results of basic information, the scale and type of goods are determined. Quantity and weight monitoring strategies are matched according to the determined scale and type of goods to obtain the actual quantity and weight of goods in real time, as well as whether there are any deviations in actual quantity or weight. When the comparison exceeds the corresponding first preset difference threshold or there is an actual quantity or weight deviation, a corresponding quantity and weight warning is triggered. The quantity and weight monitoring strategies include: single or combined monitoring methods among goods counting and weight monitoring, goods physical condition monitoring, and goods transportation route monitoring. Different goods scales and types are matched with pre-set quantity and weight monitoring strategies that are suitable for them. The goods counting and weight monitoring adopts a comprehensive monitoring method using sensors and visual recognition devices. The goods physical condition monitoring adopts a monitoring method using a visual recognition device that integrates deep learning algorithms to identify whether the goods are damaged or deformed. The goods transportation route monitoring adopts a monitoring method using visual recognition devices to monitor whether there are any missing goods along the goods transportation route.

4. The port cargo handling early warning method according to claim 1, characterized in that, Also includes: By combining real-time data, basic information monitoring results, and operational efficiency monitoring results from the target port's tallying data platform, resource allocation monitoring is performed. This includes: acquiring real-time resource allocation plans, which are pre-generated resource allocation schemes based on LSTM neural networks to predict future cargo arrivals, equipment load, personnel attendance rates, and operational efficiency; analyzing whether the real-time resource allocation plans meet the resource allocation requirements under the conditions of cargo arrivals, equipment load, personnel attendance rates, and operational efficiency obtained from the target port's tallying data platform; if the requirements are not met, a resource allocation warning is generated, and a resource allocation plan that meets the current resource allocation requirements is queried from the historical resource allocation plan database based on the acquired cargo arrivals, equipment load, personnel attendance rates, and operational efficiency, and the resource allocation plan is adjusted according to the query results. By combining real-time data from the target port's tallying data platform with the basic information monitoring process and operational efficiency monitoring process, safety and compliance monitoring is conducted. This includes: acquiring video stream data collected during the basic information monitoring and operational efficiency monitoring processes; analyzing violations in the video stream data using image analysis technology; generating safety and compliance warnings; recording the safety and compliance recovery process; marking the current basic information monitoring; and after safety and compliance recovery, re-collecting the basic information of the cargo and performing consistency comparisons. Additionally, marking the current operational efficiency monitoring, and eliminating operational delays caused by the safety and compliance recovery process during real-time monitoring of the time consumed in a single tallying operation, the unprocessed progress of batch tallying operations, and the timeliness of the tallying operation process connections.

5. A port cargo handling early warning system, characterized in that, include: The cargo handling data platform construction module is used to build a cargo handling data platform for the target port, and dynamically integrate and update customs declaration data, cargo owner bill of lading data, cargo-related document data, port equipment data, port operation personnel data, and port weather data associated with the target port. The cargo handling basic information early warning module is used to monitor basic information by combining real-time data from the target port's cargo handling data platform. This includes real-time collection of basic cargo information, such as cargo identity information, ownership and destination information, document information, cargo size and type, and comparing it with the basic information recorded in the customs declaration data, cargo owner's bill of lading data, and cargo-related document data in the target port's cargo handling data platform. If the consistency comparison fails, the corresponding basic information early warning is triggered. The cargo handling operation efficiency early warning module is used to monitor operation efficiency by combining real-time data and basic information monitoring results from the target port's cargo handling data platform. This includes real-time monitoring of the time consumed by a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling operation process connections, and comparing these with corresponding preset benchmark values. If the comparison shows a value exceeding the preset benchmark value, a corresponding operation efficiency early warning is triggered. The preset benchmark values ​​are obtained by querying historical data from the target port, under conditions of consistent cargo size and type comparison using the same port equipment data, port personnel data, port meteorological data, and basic information. The operation efficiency monitoring includes: Predefine contingency events and triggering conditions, and determine the directly and indirectly affected stages of tallying operations for each contingency event; the contingency events include pre-operation contingency events, during-operation contingency events, and post-operation contingency events; pre-operation contingency events include regional yard oversaturation, vessel arrival delays, and temporary changes in policies or regulations; during-operation contingency events include sudden cargo damage and sudden environmental disturbances; post-operation contingency events include cargo delays and yard congestion, and external supply chain disruptions; The system monitors in real time whether an emergency event is triggered. Upon detection of an emergency event, it identifies the directly affected and indirectly affected stages of the cargo handling operation. For the directly affected stages, it determines the extended operation time based on the emergency event and generates corresponding stage operation time adjustment durations to adjust the real-time monitoring of the time consumed by a single cargo handling operation and the unprocessed progress of batch cargo handling operations. For the indirectly affected stages, it matches a corresponding emergency delay prediction model based on the identified emergency event type. It determines the extended operation time based on the matched emergency delay prediction model and generates corresponding stage operation time adjustment durations to adjust the real-time monitoring of the time consumed by a single cargo handling operation and the unprocessed progress of batch cargo handling operations. Several emergency delay prediction models exist, each adapting to a type of emergency event. Different emergency delay prediction models employ deep learning algorithms with different structures or parameters and are all trained and generated using historical operation extension durations corresponding to the corresponding types of emergencies. The cargo handling quantity and weight early warning module is used to monitor the quantity and weight by combining real-time data and basic information monitoring results from the target port's cargo handling data platform. It obtains the actual quantity and weight of the cargo in real time and calculates the difference between the actual quantity and weight and the quantity and weight obtained from customs declaration data, cargo owner's bill of lading data, and cargo-related document data. The difference is then compared with a corresponding first preset difference threshold. If the comparison exceeds the corresponding first preset difference threshold, a corresponding quantity and weight early warning is triggered. The first preset difference threshold is determined by comprehensively calculating the historical quantity or weight difference threshold impact value obtained by querying cargo types that have passed consistent comparison in the same port meteorological data and basic information, and the preset basic difference threshold. The multi-port cargo handling early warning optimization module is used to construct a multi-target port cargo handling data exchange platform, determine the relationships between multiple target ports, including transshipment relationships and distribution relationships; during the basic information monitoring of multi-target port cargo handling for transshipment relationships, a consistency comparison is performed on the basic information of goods collected in real time for each port; when the consistency comparison fails, a corresponding basic information early warning is triggered; when monitoring the quantity and weight of multi-target port cargo handling for transshipment relationships, it queries whether the difference between the actual quantity and weight of goods obtained in real time from any two target ports exceeds the corresponding second preset difference threshold. When the difference between the actual quantity and weight of goods obtained in real time from any two target ports exceeds the corresponding second preset difference threshold, a corresponding quantity and weight warning is triggered. When monitoring the operational efficiency of cargo handling at multiple target ports with transshipment relationships, it checks whether at least two target ports have a preset benchmark similarity greater than a preset similarity. If found, it compares the time consumed in a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connections between any two target ports. If the difference exceeds the preset difference, a corresponding operational efficiency warning is triggered. For cargo handling at multiple target ports with distribution relationships, a baseline similarity is used. During the basic information monitoring process, for each port, the identity information, ownership, and destination information of the goods are collected in real time and compared for consistency. Document information is also collected in real time, and the size and type of the goods are combined according to the distribution relationship and then compared for consistency. If the consistency comparison fails, a corresponding basic information warning is triggered. When monitoring the quantity and weight of cargo at multiple target ports with distribution relationships, it checks whether there are target ports with distribution relationships. The difference between the actual quantity and weight of the goods obtained in real time and the difference after combining them according to the distribution relationship is greater than the corresponding second preset difference threshold. If a distribution relationship is found... When the difference between the actual quantity and weight of goods obtained in real time from the target port of the relationship and the value after being merged according to the distribution relationship is greater than the corresponding second preset difference threshold, the corresponding quantity and weight warning is triggered; when monitoring the operation efficiency of cargo handling at multiple target ports with distribution relationship, it is checked whether there are at least two target ports with a preset benchmark similarity greater than the preset similarity. When the query is found, the time consumption of a single cargo handling operation, the unprocessed progress of batch cargo handling operations, and the timeliness of cargo handling process connection are compared after the distribution relationship is merged. If the difference is greater than the preset difference, the corresponding operation efficiency warning is triggered; The multi-port tallying early warning optimization module is also used to construct a multi-target port tallying data exchange platform, determine the process of correlation between multiple target ports, sort out the order of transshipment time in transshipment correlation and the order of distribution in distribution correlation; according to the order of transshipment or distribution correlation, in the process of monitoring basic information, operational efficiency and quantity and count for the previous target port, once there is a basic information monitoring early warning, operational efficiency early warning and quantity and count early warning for the previous target port, the corresponding port tallying data is marked, and the monitoring frequency is increased in the process of monitoring basic information, operational efficiency and quantity and count for the next target port.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the method as described in any one of claims 1 to 4.

7. A computer device, characterized in that, The computer device includes a memory, a processor, and a program stored in and executable on the memory, the program being executed by the processor to implement the steps of the method as described in any one of claims 1 to 4.

Images