Warehouse management system for bulk wharf

Abstract

The invention relates to the technical field of wharf warehouse management, and discloses a warehouse management system for a bulk wharf, and the system comprises a data collection and sensing module which collects and standardizes field operation data; the data analysis and decision support module is used for analyzing historical operation data to generate a dynamic credit score; the storage and storage yard management module is used for maintaining digital twinborn storage yards and calculating candidate storage yard positions; the intelligent reservation and auditing module executes multi-dimensional auditing and generates a digital reservation voucher; the logistics execution and scheduling module is used for verifying the identity, planning the optimal path and comparing the weight; and the man-machine interaction and display module is used for generating a visual interface of storage yard stock and vehicles. Based on the preset cargo chemical compatibility matrix and the real-time environment parameters, the candidate stacking position list of the safety isolation and environment requirements is calculated and audited, and complex safety specifications are internalized into decision logic executed by the system, so that the probability of occurrence of safety accidents such as cargo cross contamination is reduced.

Description

A warehouse management system for bulk cargo terminals Technical Field

[0001] This application relates to the field of terminal storage management technology, specifically a storage management system for bulk cargo terminals. Background Technology

[0002] Bulk terminals, as core hubs in the bulk commodity supply chain, undertake the loading, unloading, storage, and transshipment of goods such as ores, coal, and grains. The efficiency, safety, and sophistication of their internal operations directly determine regional logistics costs and supply chain stability. Therefore, optimizing terminal warehousing management and improving resource turnover and operational safety are ongoing technical challenges in this field.

[0003] To improve management efficiency, some information management systems have been implemented in the industry. These systems typically achieve digital ledger management of warehouse inventory, recording basic information on goods entering and leaving the warehouse. Some systems offer online reservation functions, allowing cargo owners to submit warehousing requests containing information such as the type of goods and estimated weight through a client application. At the logistics execution level, some terminals have deployed license plate recognition equipment at entrances and exits to verify the identity of reserved vehicles, thereby achieving automated gate access control.

[0004] However, existing terminal storage management technologies still heavily rely on manual experience for yard allocation decisions. Dispatchers struggle to consistently balance the complex chemical compatibility of goods with real-time changing environmental conditions when allocating storage spaces, creating potential risks of cross-contamination, cargo damage, and even safety incidents. Customer credit management is also rather rudimentary, with existing evaluation systems primarily based on post-incident penalties for violations, lacking objective quantitative assessments of operational behaviors (such as punctuality and information accuracy) and failing to establish positive incentive mechanisms for trustworthy customers. Furthermore, in-yard logistics scheduling strategies are rigid, and vehicle route guidance is typically static, making dynamic adjustments based on real-time congestion levels difficult, resulting in low traffic efficiency. Therefore, this invention provides a storage management system for bulk cargo terminals to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this application is to provide a warehouse management system for bulk terminals, which solves the problems that existing bulk terminal warehouse management models typically suffer from: limited data collection dimensions, isolated data across different operational stages, reliance on manual experience for yard allocation and route planning, and highly subjective methods for assessing customer credit.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a bulk cargo terminal storage management system, comprising: a data acquisition and sensing module, used to receive and standardize data streams from the operational site to generate standardized data objects containing vehicle identification information, cargo weighing information, vehicle location coordinates within the yard, and environmental parameter information; a data analysis and decision support module, used to aggregate and analyze historical operational data, including historical cargo weighing information and vehicle location coordinates within the yard, to generate dynamic credit scores using a quantitative evaluation model; and a storage and yard management module, used to maintain a digital twin model containing yard inventory status data and to calculate candidate yards based on real-time environmental parameter information. The system includes a candidate storage location list; an intelligent reservation and approval module, which performs multi-dimensional constraint approval on storage reservation requests based on the calculated candidate storage location list and the dynamic credit score, and generates a digital reservation voucher that binds the vehicle identification information and cargo attributes after the multi-dimensional constraint approval is passed; a logistics execution and scheduling module, which compares the digital reservation voucher with the vehicle identification information to verify the vehicle's identity, and uses a comprehensive cost function that evaluates the degree of congestion based on the vehicle's location coordinates in the storage yard to generate the optimal route, and compares the cargo weighing information with the reservation data in the digital reservation voucher; and a human-computer interaction and display module, which generates a visual display interface by rendering the storage yard inventory status data and the vehicle's location coordinates in the storage yard.

[0007] Preferably, the step of the intelligent reservation and review module performing multi-dimensional constraint review further includes: when the candidate storage location list is not empty, determining that the safety and environmental compliance verification is passed; when the dynamic credit score is not lower than the preset credit score threshold, determining that the cargo owner's credit status verification is passed; after both the safety and environmental compliance verification and the cargo owner's credit status verification are passed, performing the final inventory capacity verification to determine whether there is a storage location in the candidate storage location list that meets the quality requirements of the reserved goods.

[0008] Preferably, the warehousing and yard management module performs multi-stage filtering decision calculations through a safety planning decision engine to generate the candidate storage location list. The steps of performing multi-stage filtering decision calculations further include: a first stage, performing safety isolation filtering, based on a preset cargo chemical compatibility matrix, selecting storage locations that do not have chemical conflicts with the cargo already stored in adjacent storage locations; a second stage, performing environmental compliance filtering, based on the real-time environmental parameter information, further selecting storage locations that meet the environmental requirements of the cargo to be stored from the storage locations selected in the first stage.

[0009] Preferably, the input parameters of the comprehensive cost function further include the physical length of the road and the safety risk coefficient.

[0010] Preferably, when the deviation between the cargo weighing information and the reservation data exceeds a preset severe over-limit threshold, the logistics execution and scheduling module triggers a closed-loop anomaly handling process, which includes suspending the operation and resubmitting it to the intelligent reservation and review module for secondary verification.

[0011] Preferably, when evaluating operational deviations, the quantitative evaluation model applies a credit score penalty if the evaluated operational deviation exceeds a preset threshold.

[0012] Preferably, the data analysis and decision support module includes: a multi-source data fusion unit, used to associate and aggregate the scattered historical operation data around a single scheduled operation to generate a complete operation history record; an operation efficiency and credit dynamic evaluation model unit, used to generate a dynamic credit score based on the complete operation history record and according to the cargo owner's punctuality, information accuracy, and operation efficiency by applying a quantitative evaluation model; and a system self-optimization strategy generation unit, used to analyze the vehicle's location coordinate information in the yard to identify normalized congestion points and generate instructions for adjusting the weight coefficients corresponding to the congestion level in the comprehensive cost function.

[0013] Preferably, the data acquisition and sensing module includes: a data interface unit, used to establish a communication link with heterogeneous data sources through a protocol adaptation layer that integrates multiple communication protocols; a data processing and standardization unit, used to parse and verify the data stream received from the work site by the data interface unit, and convert it into a predefined unified data model to generate the standardized data object; and a data distribution and buffering unit, used to publish the standardized data object to the message middleware inside the warehouse management system using a publish-subscribe model.

[0014] Preferably, the human-computer interaction and display module includes a role-based view generation unit, which dynamically constructs and presents a user interface and function set that matches the role based on the logged-in user's role identifier.

[0015] Preferably, the visualization interface dynamically loads and renders the following data layers on top of the digital map base map: an inventory status layer, including yard inventory status data obtained from the warehousing and yard management module; and a vehicle dynamic layer, including the vehicle's location coordinates within the yard.

[0016] In summary, this application includes at least one of the following beneficial technical effects: 1. By introducing multi-dimensional constraints in the reservation and review stage and performing multi-stage filtering in the yard allocation stage, the present invention specifically calculates a list of candidate storage locations that meet safety isolation and environmental requirements based on a preset cargo chemical compatibility matrix and real-time environmental parameters. The intelligent reservation and review module reviews this list as a prerequisite. By internalizing complex safety regulations into a deterministic decision-making logic that is automatically executed by the system, the probability of safety accidents such as cross-contamination and chemical reactions caused by human planning negligence or lack of experience is reduced.

[0017] 2. The data analysis and decision support module of this invention converts cargo owners' historical operational data (such as punctuality, information accuracy, and operational efficiency) into dynamic credit scores using a quantitative evaluation model. The intelligent reservation and approval module directly applies this score as a key judgment criterion in the approval process. This mechanism replaces traditional subjective evaluations or simple records of violations with objective data, making the credit assessment results more equitable. Furthermore, by linking reservation permissions, it positively incentivizes cargo owners to regulate their operational behavior, thereby improving the overall operational order and collaborative efficiency of the terminal.

[0018] 3. The comprehensive cost function adopted by the logistics execution and scheduling module of this invention comprehensively considers the physical length of the road, the degree of congestion assessed based on real-time vehicle location coordinates, and the safety risk coefficient determined by the cargo operation status when performing dynamic path planning. This ensures that the generated path balances efficiency and safety, reduces the ineffective waiting time of vehicles in the yard, and the data analysis and decision support module identifies common congestion points by analyzing historical data and can generate instructions to adjust the congestion weight coefficient. This forms a closed loop of data analysis, parameter optimization, and scheduling execution, enabling the system to learn and evolve on its own and continuously optimize its scheduling capabilities. Attached Figure Description

[0019] Figure 1 is a system architecture diagram of this application; Figure 2 is a flowchart of the intelligent reservation and multi-dimensional review process of this application; Figure 3 is a flowchart of the vehicle entry operation and dynamic scheduling process of this application. Detailed Implementation

[0020] The present application will be further described in detail below with reference to Figures 1-3.

[0021] Referring to Figure 1, Figure 1 is a system architecture diagram according to an embodiment of the present invention. The present invention provides a bulk terminal warehousing management system, including a data acquisition and sensing module, an intelligent reservation and approval module, a warehousing and yard management module, a logistics execution and scheduling module, a data analysis and decision support module, and a human-computer interaction and display module.

[0022] The data acquisition and sensing module receives data streams from various front-end devices or information systems at the dock operation site through standardized data interfaces. These data streams include, but are not limited to, vehicle identification information, cargo weighing information, vehicle location coordinates within the yard, and environmental parameters such as wind speed, wind direction, and humidity. This module parses, cleans, and standardizes the received heterogeneous data, converting it into a structured data stream in a unified format for use by other modules within the system.

[0023] The intelligent reservation and approval module processes warehousing reservation requests initiated by cargo owners via user terminals. Upon receiving a reservation application containing information such as cargo attributes, estimated weight, and planned arrival time, it collaboratively calls interfaces from other modules to perform multi-dimensional constraint approval. This approval process not only verifies basic inventory capacity but also includes a safety compatibility assessment of the goods to be stored and those already in the yard, a compliance assessment of the operational environment conditions, and a credit assessment of the cargo owner's historical operational behavior. After approval, the module generates a globally unique digital reservation voucher binding all reservation information and distributes it to relevant parties.

[0024] The warehousing and yard management module is used to build and maintain a digital twin model of the bulk terminal yard. This module records and updates the inventory status of storage locations in real time, including total capacity, used capacity, and available capacity. A dynamic yard planning and security isolation unit is deployed within the module, containing a cargo attribute knowledge base and a security compatibility rule set. Upon receiving a reservation approval request, based on the physicochemical properties of the goods to be received and combined with real-time environmental data obtained from the data acquisition and sensing module, a list of candidate storage locations that meet security isolation and environmental protection requirements is calculated and generated for the intelligent reservation and approval module to make a decision.

[0025] The logistics execution and scheduling module is activated after vehicles enter the facility after automatic identity verification. It provides end-to-end operational guidance for vehicles within the facility. Its internal dynamic path planning unit generates the optimal driving route for vehicles based on a comprehensive cost function. The input parameters for this function include real-time traffic conditions, route length, and the safety risk level of the areas traversed. During operations, the weighing comparison and anomaly handling unit of this logistics execution and scheduling module compares the actual weighing data with the reservation data. If a significant discrepancy occurs, it triggers a closed-loop feedback process to the intelligent reservation and verification module for secondary verification of warehouse capacity.

[0026] The data analysis and decision support module periodically collects and integrates all historical data generated by various modules within the system during the operational process through a multi-source data fusion unit. Its internal dynamic operational efficiency and credit evaluation model, based on a pre-set quantitative evaluation algorithm, calculates dynamic credit scores for cargo owners' operational behaviors, such as punctuality, information accuracy, and operational efficiency. Furthermore, through in-depth analysis of historical data, it can generate system optimization strategy parameters, such as adjusting the weighting coefficients in the route planning function to achieve adaptive adjustments to the system's scheduling capabilities.

[0027] The human-computer interaction and display module is used to provide users with different roles, such as cargo owners, drivers, and terminal managers, with functional views and data access interfaces that match their permissions. It is also used to provide a visual display of the yard. It calls data from the warehousing and yard management module to render the inventory status of each yard, the dynamic location of vehicles in the yard, and the operating status of operating equipment on a 2D or 3D digital map in real time. This human-computer interaction and display module is also responsible for pushing information such as system-generated work instructions, review results, and abnormal alarms to the designated user terminals in real time and accurately.

[0028] In a specific implementation, the data acquisition and sensing module includes a data interface unit, a data processing and standardization unit, and a data distribution and buffering unit.

[0029] The data interface unit is responsible for establishing communication links with various heterogeneous data sources deployed at the dock operation site. To adapt to data from different sources, this data interface unit integrates a protocol adaptation layer, including implementations of multiple communication protocols. For IoT devices such as high-precision positioning equipment or environmental sensors installed on stacker-reclaimers and belt conveyors, data subscription is performed using the lightweight Message Queuing Telemetry Transport Protocol (MQTT). For license plate recognition cameras or some modern weighbridge systems, this unit uses Hypertext Transfer Protocol (HTTP / HTTPS) to call their provided application programming interface (RESTful API) to obtain recognition results or weighing data. For some traditional devices that do not provide network interfaces, it supports monitoring the underlying data stream through Transmission Control Protocol (TCP / IP) sockets or monitoring changes to data files in a specified directory through file system polling. This provides a wide range of access capabilities for multi-source heterogeneous data.

[0030] The data processing and standardization unit receives raw, heterogeneous data streams from the data interface unit and performs a series of processes to output unified, standardized internal data objects. This processing can be formally expressed by the following function: D std =F proc (D raw ); where D rawD represents the raw data points received from an external data source. std This represents the standardized data object generated after processing, while F proc This indicates a composite processing function, which internally includes at least the following steps: a data parsing step, based on D... raw The source and format of the data (such as JSON, XML, or a custom binary format) are selected by an appropriate parser to extract the content into a temporary data structure in memory. The data validation step verifies the validity of the parsed data according to a pre-defined set of validation rules. The validation rule set includes, but is not limited to: numerical range checks (e.g., cargo weight must be positive), format checks (e.g., license plate numbers must conform to a specific regular expression), and integrity checks (e.g., key fields cannot be empty). Data that fails validation is marked as invalid and stored in the exception log, and will not proceed to subsequent processes. The data standardization step converts valid data that passes validation into a predefined unified data model within the system. For example, regardless of the original data format, all time-related information is converted to a Coordinated Universal Time (UTC) timestamp in ISO 8601 format; all geographic location information is converted to latitude and longitude coordinates in the WGS-84 coordinate system. Through this step, all data objects entering the system have a consistent structure and semantics. For example, a weighing data object can be uniformly represented as a structured object containing standard fields such as device ID, timestamp, data type (e.g., weight), value, and unit, thereby eliminating ambiguity when upper-level modules process data.

[0031] The data distribution and buffering unit is responsible for reliably and efficiently pushing standardized data objects generated by the data processing and standardization unit to other functional modules within the system that require this data. In this embodiment, a publish-subscribe design pattern is adopted. Whenever a standardized data object is generated, this unit publishes it to a specific topic on the system's internal message bus or message middleware (such as Kafka or RabbitMQ). For example, weighing data is published to "topic.measurement.weight", and license plate recognition data is published to "topic.vehicle.lpr". Other modules within the system, such as the logistics execution and scheduling module or the data analysis and decision support module, can subscribe to topics of interest as needed. This implementation not only decouples the data acquisition module from the data consumption module, improving the system's flexibility and scalability, but also provides a buffering mechanism for the data stream by utilizing the queuing and persistence capabilities of the message middleware. This ensures that data is not lost in the event of a brief unavailability or slow processing of downstream modules, thereby enhancing the robustness of the entire system.

[0032] Referring to Figure 2, in a specific implementation, the intelligent appointment and approval module includes an appointment information receiving and parsing unit, a multi-dimensional constraint approval unit, and an appointment voucher generation and distribution unit.

[0033] The reservation information receiving and parsing unit receives standardized reservation request data objects from the human-computer interaction and display module. This data object contains structured reservation information fields, such as: shipper identification, transport vehicle license plate, cargo type code, estimated cargo quality, and planned arrival time window. Upon receiving the data object, this unit first performs format and completeness checks to ensure that all required fields exist and conform to preset data type and format specifications. Then, it encapsulates the validated reservation request information into an internal processing object and passes it to the multi-dimensional constraint review unit.

[0034] The multi-dimensional constraint verification unit, upon receiving an internal processing object, initiates a parallel or sequential multi-dimensional constraint verification process. The final result of this process is A. result It can be formally represented as a logical AND operation: A result =F sec ∧F credit ∧F inv Among them: A result The final audit result is a Boolean value, and is true if and only if all validation functions return true.

[0035] F sec This represents a safety and environmental compliance verification function. To perform this verification, information such as the reserved cargo type code is sent via an internal interface to the dynamic yard planning and safety isolation unit in the warehousing and yard management module. This call aims to obtain a "list of available storage locations" that currently meet safety isolation and environmental requirements. If the returned list is empty, it means that there are no storage locations that meet safety specifications available for allocation under the current yard conditions. sec The result is false.

[0036] F credit This represents the function for verifying the shipper's credit status. This multi-dimensional constraint verification unit identifies the scheduled shipper and calls the operational efficiency and credit dynamic evaluation model in the data analysis and decision support module via an internal interface. This call aims to obtain the shipper's current real-time credit score, compare it with a system-preset credit score threshold, and only if the score is not lower than the threshold will F... credit The result is true.

[0037] F inv This represents the final inventory capacity verification function. This function is executed only if the results of the first two verification functions are both true. It iterates through F... secThe process returns a "List of Available Storage Locations," which is then checked one by one to ensure that the real-time available capacity of each storage location is not less than the estimated weight of the goods to be booked. If a storage location that meets the capacity requirements is found in the list, then F... inv If the result is true, the stack location will be used as the proposed stack location for this reservation.

[0038] When A result If true, the unit will submit a reservation request with the proposed stacking location information to the manual review queue, or automatically approve it according to the system configuration. If A result If false, the unit generates a rejection response containing a specific failure reason code (such as "security conflict", "insufficient credit" or "insufficient capacity") and returns it.

[0039] The reservation voucher generation and distribution unit is activated after the reservation request is finally approved. It generates a globally unique reservation identifier (e.g., a universally unique identifier (UUID) or a hash value generated based on timestamps and business information). Subsequently, in the database transaction, this reservation voucher generation and distribution unit strongly associates this unique identifier with all the core information of this reservation. This core information includes the shipper's identity, vehicle license plate, cargo type, estimated weight, planned time, and the final determined storage location number, thus forming a complete and indivisible digital operation voucher. After the voucher is generated, the message push service interface provided by the human-computer interaction and display module is invoked to accurately distribute the reservation voucher (which can be represented as a QR code image, barcode, or plain text encoding) along with brief operation information to the user terminals of the shipper and driver associated with the reservation via mobile application push, SMS, or email.

[0040] In one specific implementation, the warehousing and yard management module includes a real-time inventory monitoring unit, a dynamic yard planning and safety isolation unit, and a warehousing operation recording unit.

[0041] The real-time inventory monitoring unit is responsible for building and maintaining a digital twin model in the system database that precisely corresponds to the physical yard of the terminal. This digital twin model, in the form of a data structure, describes the entire yard as a collection of storage location objects. Each storage location object contains a series of static attributes (such as storage location number, geographical coordinates, total design capacity, and load-bearing capacity) and dynamic attributes (such as currently used capacity, remaining available capacity, currently stored cargo type code, and occupancy status). The real-time inventory monitoring unit subscribes to precise inbound and outbound event messages published by the logistics execution and scheduling module. When a precise inbound event message is received, based on the actual inbound cargo quality contained in the message, the "used capacity" attribute value of the corresponding storage location object is increased atomically, and its "remaining available capacity" attribute value is decreased accordingly. Outbound operations are performed in the opposite manner. This event-driven update mechanism ensures the real-time nature and accuracy of the inventory data in the digital twin model.

[0042] The dynamic yard planning and safety isolation unit is used to intelligently select candidate storage locations that meet multiple safety and environmental constraints for goods to be stored during the reservation phase. It mainly relies on a cargo attribute knowledge base and a safety planning decision engine.

[0043] The cargo attribute knowledge base is a structured database or key-value storage system that stores detailed attribute information for various types of bulk cargo. Each type of cargo is indexed by its unique cargo type code. The stored attribute fields include at least: physical state (e.g., powder, granules, lumps), chemical properties (e.g., oxidizing power, corrosiveness, flammability), dust explosion index (Kst value), whether it is a dangerous good and its corresponding UN Dangerous Goods Number (UN Number), and storage environment requirements (e.g., rainproof, moisture-proof, light-proof).

[0044] The safety planning decision engine, when invoked by the intelligent reservation and approval module, receives the type code of the goods to be received and the quality of the reserved goods as input, and executes a multi-stage filtering decision process to output a list of qualified candidate storage locations. This process can be described as follows: The first stage involves safety isolation filtering. The decision engine first obtains a snapshot of the current state of the entire storage yard from the real-time inventory monitoring unit. For non-empty storage locations, the engine queries the cargo attribute knowledge base to obtain the chemical properties of the stored goods. Subsequently, the engine traverses all idle or remaining storage locations and checks whether its stored goods meet the safety isolation requirements with those of all adjacent storage locations (adjacency relationships are defined by a preset storage yard topology). This determination is based on a preset cargo chemical compatibility matrix M. chem This matrix is ​​an N×N Boolean matrix, where N is the total number of goods types in the knowledge base. If element M in the matrix... chem A value of 0 for [i,j] indicates that goods type i and goods type j cannot be stored adjacently. Stack position s can only be used if it is adjacent to all its neighboring stack positions s. ′The existing goods g ′ All satisfy M chem [g req ,g ′ Under the condition that ] = 1 (where g req Only goods of the type to be received into the warehouse can pass this stage of filtering.

[0045] The second stage involves environmental compliance filtering. For the set of locations that passed the first stage filtering, the decision engine further calls the interface of the data acquisition and sensing module to obtain real-time environmental parameters, such as wind direction D. wind And rainfall conditions. The engine performs secondary screening based on the environmental requirements of the goods to be stored. For example, if the goods to be stored are marked as "requires rain protection", all open-air storage locations will be removed from the candidate list when rainfall is detected. As another example, if the goods to be stored are materials that are prone to generating dust, and the real-time wind direction is westerly, all storage locations on the east side of the storage yard that are adjacent to dust-sensitive goods (such as grain) will also be removed from the candidate list to avoid cross-contamination.

[0046] The third stage involves capacity and load-bearing capacity filtering. For the final candidate stacking locations that have passed the filtering in the first two stages, the decision engine checks whether the "remaining available capacity" of each stacking location is not less than the reserved cargo weight, and whether the "load-bearing capacity" of the stacking location meets the requirements.

[0047] The safety planning decision engine returns the stacking list filtered through all three stages as the result to the caller, internalizing complex safety specifications and dynamic environmental factors into a deterministic decision-making logic that is automatically executed by the system, thereby preventing safety accidents and cargo losses caused by human planning negligence at the source.

[0048] The warehousing operations recording unit is responsible for accurately tracking the lifecycle of goods within the warehouse. When a batch of goods is confirmed to be received, this unit creates a warehousing record in the database, which includes goods information, the location of the goods upon receipt, the exact timestamp of receipt, and the planned warehousing duration set during the reservation. By running a scheduled task, all warehousing records in the warehouse are periodically scanned, and the current time is compared with the timestamp of receipt to calculate the actual warehousing duration. When the actual warehousing duration exceeds the planned warehousing duration and the preset grace period, an overdue alarm event is automatically generated and published to the system message bus to notify relevant management personnel and trigger potential overdue warehousing billing procedures.

[0049] Referring to Figure 3, in a specific implementation, the logistics execution and scheduling module includes a vehicle authentication unit, a route dynamic planning unit, and a weighing comparison and anomaly handling unit.

[0050] The vehicle authentication unit is the starting point for automating the vehicle entry process. This unit subscribes to vehicle identification information topics published by the data acquisition and sensing module. When a vehicle arrives at the terminal entrance, the data acquisition and sensing module captures its license plate image, identifies the license plate number, and publishes it as a standardized data object. Upon receiving this data object, the vehicle authentication unit immediately queries the database of valid reservation vouchers generated by the intelligent reservation and verification module, using the license plate number as an index. The query logic includes verifying whether the license plate exists in the valid reservation list for the day, and verifying whether the current time is within the planned entry time window bound to the reservation voucher (which may include a preset grace period). If the query is successful, the unit sends a "allow passage" command to the terminal gate control system; if the query fails, an entry failure event is generated, and the reason for the failure (such as "no reservation" or "not within the reserved time period") is pushed to the terminal management personnel's terminal.

[0051] The path dynamic planning unit is activated after a vehicle has been authenticated and entered the yard. It plans the optimal route from the vehicle's current location to a designated work point (such as a weighing area or unloading stack). This unit first loads the internal road network topology map of the terminal from the digital twin model maintained by the warehousing and yard management module. This topology map is implemented as a weighted directed graph G = (V, E), where the set of nodes V represents intersections or key locations, and the set of edges E represents road segments connecting these nodes. Each edge e ∈ E is associated with a dynamically changing travel cost C(e). An improved Dijkstra's algorithm or A* is used to find the shortest weighted path between the starting and ending points. The travel cost C(e) is calculated as a function incorporating multiple dynamic factors: C(e) = w d ·L(e)+w c ·T cong (e)+w s ·R safe (e); where: C(e) is the comprehensive toll cost of edge e (i.e., road segment); L(e) is the physical length of road segment e, which is a static value; T cong (e) represents the predicted congestion time for road segment e. This is dynamically calculated by the congestion model, which continuously analyzes real-time vehicle location data subscribed to from the data acquisition and sensing module, and assesses the degree of congestion by statistically analyzing vehicle density or average driving speed on the road segment.

[0052] R safe (e) is the safety risk coefficient for road segment e. This value is linked to the storage and yard management module. For example, when the path dynamic planning unit detects that road segment e is adjacent to a storage location where dust material unloading is underway, it will obtain the safety impact radius of the operation from the storage and yard management module and increase R accordingly. safe The value of (e) is used to guide other vehicles away from the potentially polluted area.

[0053] w d ,w c ,w s These are preset weighting coefficients for distance, congestion, and safety risks, which port management personnel can adjust according to different time periods (such as prioritizing traffic efficiency during peak hours and prioritizing safety at night).

[0054] After calculating the optimal route, it is broken down into specific navigation instructions (such as "turn right at the intersection 100 meters ahead"), and pushed to the driver's in-vehicle or handheld terminal in real time through the human-computer interaction and display module.

[0055] The weighing comparison and anomaly handling unit is responsible for automatically verifying and processing cargo information during the weighing process. This unit subscribes to weighing data generated by the weighbridge system. When the vehicle completes weighing, it receives a data set containing the actual weight W. actual After obtaining the data object, immediately retrieve the estimated weight W associated with the vehicle from the reservation voucher database. plan The two weight values ​​are compared, and different business logic is executed based on the comparison result. If the weight deviation is within the preset reasonable error range, i.e., |W actual -W plan | / W plan If the error is less than or equal to ∈ (where ∈ is the allowed percentage of error), the process continues normally, the system updates the navigation instructions, and guides the vehicle to the final unloading location.

[0056] If the weight deviation exceeds the reasonable error range but does not reach the threshold of serious over-limit, the system can automatically calculate the supplementary price or refund according to the preset business rules and record the financial adjustment.

[0057] When the weight deviation exceeds a more stringent severe over-limit threshold, namely |W actual -W plan |>δ w,max Upon receiving this event, a closed-loop exception handling and secondary decision-making process will be triggered. Subsequent operations for the vehicle will be immediately suspended, and a "major anomaly" event will be constructed. This event, along with all appointment information and actual weighing data, will be resubmitted to the intelligent appointment and review module via an internal interface. Upon receiving this anomaly event, the intelligent appointment and review module will use W... actual As input, the warehousing and yard management module is invoked again to perform a second verification of the remaining capacity and load-bearing capacity of the originally designated storage location. If the second verification passes, the process resumes; if it fails, the intelligent reservation and approval module decides to implement a backup plan, such as reallocating the storage location or canceling the current operation. This closed-loop mechanism ensures that the system's decisions are dynamically corrected based on the most accurate on-site data, avoiding potential safety risks or operational failures.

[0058] In one specific implementation, the data analysis and decision support module includes a multi-source data fusion unit, an operational efficiency and credit dynamic evaluation model unit, and a system self-optimization strategy generation unit.

[0059] The multi-source data fusion unit is responsible for providing high-quality, well-organized datasets for analysis and modeling. Through a configurable Extract, Transform, Load (ETL) process, it periodically aggregates various business data from databases or message buses of other modules within the system. This process correlates and aggregates data scattered across different business stages around a single scheduled operation. For example, it aggregates the planned arrival time and estimated cargo quality (from the intelligent scheduling and approval module) with actual arrival time, actual weighing quality, total on-site operation time (from the logistics execution and scheduling module), and whether overdue storage alarms were triggered (from the storage and yard management module), into a complete operation history. During the aggregation process, the data is also cleaned and transformed to ensure data consistency and availability.

[0060] The operational efficiency and credit dynamic assessment model unit incorporates a quantitative assessment model to dynamically calculate the shipper's credit score based on aggregated operational history. This quantitative assessment model transforms the evaluation of customer business credit from subjective judgment to objective, data-driven decision-making. The specific credit score calculation formula is expressed as follows: Wherein: S credit This represents the final credit score calculated for the cargo owner at the end of the assessment period. This score is stored and linked to the cargo owner's identity, serving as a crucial basis for verification during their next booking; S base This represents the shipper's initial base score, which is usually set to a fixed positive integer, such as 100; w t ,w w ,w e These represent the weighting coefficients of time deviation, weight deviation, and operational efficiency in the overall credit assessment system. These coefficient values ​​can be configured by terminal managers according to operational strategies to adjust the impact of different behaviors on the credit score; f t (·) is the time deviation penalty function. In a specific implementation, this function is designed as a piecewise function. When the time difference (T) actual -T plan Within the preset grace period, the function value is 0; after the grace period, the function value monotonically increases with the delay time, for example, it can be a linear or exponential growth function, thus realizing the quantitative penalty for lateness; w(·) is the weight deviation penalty function. This function takes the relative percentage deviation of the weight as input. In a specific implementation, when the absolute value of the relative deviation is less than a preset exemption threshold (e.g., 2%), the function value is 0; when it exceeds the threshold, the function value monotonically increases with the percentage deviation, thereby achieving a quantitative penalty for inaccurate cargo information. This represents the total penalty score for all recorded violations within the assessment period. A violation penalty table is maintained, where different types of violations (P) are represented. i For example, unjustified cancellation of appointments, exceeding the weighing limit and triggering a secondary decision, or falsely reporting dangerous goods, different fixed deduction values ​​are defined. vio f is the total number of violations during the period. e (·) is the operational efficiency reward function. This function aims to incentivize cargo owners and their partner fleets to improve operational efficiency on-site; T op T represents the total actual time taken from the vehicle's entry to the completion of unloading and departure. bench This refers to the historical average processing time or the benchmark processing time set by management for similar goods and operating conditions (such as similar transportation distances). When the ratio T... bench / T op When the ratio is greater than 1 (i.e., the actual time taken is less than the baseline time), the function outputs a positive bonus score; when the ratio is less than 1, the function value is 0 or a small negative value.

[0061] The system's self-optimization strategy generation unit automatically generates strategy suggestions for adjusting the operating parameters of other modules of the system by conducting in-depth analysis and pattern mining on a large amount of historical data accumulated in the data analysis and decision support module.

[0062] In one specific implementation, the system's self-optimizing strategy generation unit continuously analyzes the average travel time of all vehicles on different road segments and at different times, thereby identifying recurring congestion points. Based on this discovery, it can automatically generate adjustment instructions to increase the congestion weight coefficient w of the corresponding road segment in the path planning cost function of the logistics execution and scheduling module during specific time periods. c This allows subsequent route planning to proactively avoid the predicted congestion.

[0063] In another implementation, by analyzing the overall appointment success rate and yard utilization rate, combined with the overall distribution of cargo owner credit scores, suggestions can be made to the administrator to adjust the credit score threshold in appointment review, in order to achieve a dynamic balance between ensuring the experience of high-credit customers and improving the overall turnover rate of the yard. These generated strategy parameters can be configured for automatic application or take effect after being pushed to the administrator for confirmation, thus forming a complete closed loop of "data-analysis-decision-feedback-optimization".

[0064] In one specific implementation, the human-computer interaction and display module includes a role-based view generation unit, a visual yard display unit, and a task and alarm push unit.

[0065] The role-based view generation unit dynamically constructs and presents a user interface and set of functions matching the logged-in user's role based on the user's authentication and permission information. Internally, the role-based view generation unit maintains a role-permission mapping table. When a user logs into the system via their terminal device (such as a personal computer, tablet, or smartphone), this unit retrieves the assigned role identifier (such as "cargo owner," "driver," or "terminal manager") after successful authentication. Based on this role identifier, the unit retrieves the role-specific interface layout, component set, and data access permissions from the backend service.

[0066] For the "cargo owner" role, the generated interface will prominently display the reservation application submission form, historical reservation record query, credit score display, and related financial settlement information. For the "driver" role, the interface is greatly simplified, mainly displaying the currently pending work tasks, a map view for receiving navigation instructions, and electronic vouchers for on-site confirmation. For the "terminal manager" role, a global management dashboard with the highest authority is presented, integrating monitoring views and control entry points for all functional modules. This approach ensures that each user can only access and operate information and functions within their scope of responsibility, guaranteeing both system security and ease of use.

[0067] The visualized yard display unit is used to intuitively present the physical space and operational status of the terminal in a graphical way. In a specific implementation, a static base map of the terminal yard is loaded in the user's browser or application based on a two-dimensional or three-dimensional graphics rendering engine. On this base map, multiple interactive data layers are dynamically loaded and rendered through real-time data interfaces with the warehousing and yard management module and the logistics execution and scheduling module. For example, the "inventory status layer" obtains real-time inventory data of all storage locations from the warehousing and yard management module and colors the corresponding areas on the map with different colors or icons according to the occupancy rate of the storage locations (e.g., idle, half full, full) or the dangerous goods level of the stored goods. By clicking on a specific storage location area, a pop-up window can display its detailed information, such as the type of goods, current inventory, and owner. Another example is the "vehicle dynamic layer," which subscribes to the real-time vehicle location coordinate topics published by the data acquisition and sensing module and accurately displays the real-time location and driving trajectory of all vehicles operating on site on the map in the form of moving icons.

[0068] The task and alarm push unit establishes a persistent WebSocket connection between the browser-based web client and the server. For mobile application clients, it integrates a client software development kit (SDK) for operating system-level push notification services (such as APNS or FCM). This task and alarm push unit serves as the unified management interface for these push channels, subscribing to various critical business events on the system's internal message bus. When other modules publish events—for example, the intelligent appointment and approval module approves an appointment, the logistics execution and scheduling module generates new navigation instructions, or the data analysis and decision support module issues a timeout alarm—the task and alarm push unit can immediately capture the event. Upon capture, it formats the event into a human-readable message based on the event content and preset push rules, selects an appropriate push channel, and sends the message precisely, within milliseconds, to the terminal device of the specific user associated with the event, thereby ensuring timely issuance of work instructions and rapid response to abnormal situations.

Claims

1. A warehouse management system for bulk cargo terminals, characterized in that, include: The data acquisition and sensing module is used to receive and standardize the data stream from the work site to generate standardized data objects containing vehicle identification information, cargo weighing information, vehicle location coordinates in the yard, and environmental parameter information; the data analysis and decision support module is used to aggregate and analyze historical operation data, including historical cargo weighing information and vehicle location coordinates in the yard, to generate dynamic credit scores by applying a quantitative evaluation model. The warehousing and yard management module maintains a digital twin model containing yard inventory status data and calculates a candidate yard location list based on real-time environmental parameters. The intelligent reservation and approval module performs multi-dimensional constraint approval on warehousing reservation requests based on the calculated candidate yard location list and the dynamic credit score. Upon successful approval, it generates a digital reservation voucher that binds the vehicle identification information to the cargo attributes. The logistics execution and scheduling module compares the digital reservation voucher with the vehicle identification information to verify vehicle identity, uses a comprehensive cost function based on the vehicle's location coordinates within the yard to assess congestion, generates the optimal route, and compares the cargo weighing information with the reservation data in the digital reservation voucher. The human-computer interaction and display module generates a visual interface by rendering the yard inventory status data and the vehicle's location coordinates within the yard.

2. The bulk cargo terminal storage management system according to claim 1, characterized in that, The steps of the intelligent reservation and review module to perform multi-dimensional constraint review further include: when the candidate storage location list is not empty, determining that the safety and environmental compliance verification is passed; when the dynamic credit score is not lower than the preset credit score threshold, determining that the cargo owner's credit status verification is passed; after both the safety and environmental compliance verification and the cargo owner's credit status verification are passed, performing the final inventory capacity verification to determine whether there is a storage location in the candidate storage location list that meets the quality requirements of the reserved goods.

3. A bulk cargo terminal storage management system according to claim 1, characterized in that, The warehousing and yard management module calculates the candidate storage location list by performing multi-stage filtering decisions through a safety planning decision engine. The steps of performing multi-stage filtering decisions further include: a first stage, performing safety isolation filtering, based on a preset cargo chemical compatibility matrix, selecting storage locations that do not have chemical conflicts with the cargo already stored in adjacent storage locations; a second stage, performing environmental compliance filtering, based on the real-time environmental parameter information, further selecting storage locations that meet the environmental requirements of the cargo to be stored from the storage locations selected in the first stage.

4. A bulk cargo terminal storage management system according to claim 1, characterized in that, The input parameters of the comprehensive cost function further include the physical length of the road and the safety risk coefficient.

5. A bulk cargo terminal storage management system according to claim 1, characterized in that, When the deviation between the cargo weighing information and the reservation data exceeds a preset severe over-limit threshold, the logistics execution and scheduling module triggers a closed-loop exception handling process. The exception handling process includes suspending the operation and resubmitting it to the intelligent reservation and review module for secondary verification.

6. A bulk cargo terminal storage management system according to claim 1, characterized in that, When evaluating operational deviations, the quantitative evaluation model applies a credit score penalty if the deviation exceeds a preset threshold.

7. A bulk cargo terminal storage management system according to claim 1, characterized in that, The data analysis and decision support module includes: a multi-source data fusion unit, used to associate and aggregate the scattered historical operation data around a single scheduled operation to generate a complete operation history record; an operation efficiency and credit dynamic evaluation model unit, used to generate a dynamic credit score based on the complete operation history record and according to the cargo owner's punctuality, information accuracy, and operation efficiency by applying a quantitative evaluation model; and a system self-optimization strategy generation unit, used to analyze the vehicle's location coordinate information in the yard to identify common congestion points and generate instructions for adjusting the weight coefficients corresponding to the degree of congestion in the comprehensive cost function.

8. A bulk cargo terminal storage management system according to claim 1, characterized in that, The data acquisition and sensing module includes: a data interface unit, used to establish a communication link with heterogeneous data sources through a protocol adaptation layer that integrates multiple communication protocols; a data processing and standardization unit, used to parse and verify the data stream received from the work site by the data interface unit, and convert it into a predefined unified data model to generate the standardized data object; and a data distribution and buffering unit, used to publish the standardized data object to the message middleware inside the warehouse management system using a publish-subscribe model.

9. A bulk cargo terminal storage management system according to claim 1, characterized in that, The human-computer interaction and display module includes a role-based view generation unit, which dynamically constructs and presents a user interface and set of functions that match the role based on the logged-in user's role identifier.

10. A bulk cargo terminal storage management system according to claim 1, characterized in that, The visualization interface dynamically loads and renders the following data layers on top of the digital map base: inventory status layer, including yard inventory status data obtained from the warehousing and yard management module; The vehicle dynamic layer includes the vehicle's position coordinates within the field.

Images