Logistics transfer center digital twin visual management and control method

Abstract

The application relates to a logistics transfer center digital twin visualization management and control method, which takes blocks in field drawing data files as analysis references to restore physical coordinates and device types of each field device, recursively traverses dynamic blocks containing visibility states, constructs device topology relationships between the field devices and generates a digital canvas; binds field device identifiers and business groups to the primitives in the digital canvas, aggregates and superimposes multi-source heterogeneous data accessed according to the bound field device identifiers to the corresponding primitives; and determines target primitives and executes a visualization response based on the binding relationship of the primitives and the device topology relationship in response to a business request for the digital canvas. The application links field physical layout and multi-source operation data on a unified canvas, improves the accuracy and universality of digital station restoration and the real-time performance of field management and control.

Description

Technical Field

[0001] This application relates to the field of logistics and supply chain technology, and in particular to a digital twin visualization control method for logistics transfer centers. Background Technology

[0002] Logistics transfer centers are core nodes in the express delivery network, undertaking key functions such as parcel sorting, consolidation, and dispatch. A single transfer center typically deploys thousands of on-site devices, including cross-belt sorters, telescopic conveyors, swing wheel sorters, side-sliding machines, and conveyor lines. Parcels flow continuously within the center along the unloading, sorting, consolidation, and loading chain. Real-time monitoring and efficient management of on-site operations directly impact the transfer timeliness and service quality of the entire network.

[0003] To grasp the on-site operational status, operators need a site map that accurately reflects the layout of equipment and can be updated in real time according to the operational status. With the increase in the number of transfer centers and the expansion of the scale of a single center, the equipment layout is becoming increasingly complex, the parcel flow chain is lengthening, and the cross-system data is becoming increasingly complex. Relying solely on manual monitoring of the on-site status is no longer sustainable. On-site management has placed higher demands on automated digital site reconstruction and unified presentation of multi-source operational data.

[0004] Existing on-site visualization methods suffer from a disconnect between the physical layout and real-time operational data. On one hand, site modeling relies either on manually drawn diagrams or on custom-coded data for individual centers. The former cannot accurately reproduce equipment layouts consistent with the actual site and requires daily manual updates, while the latter is difficult to replicate and extend to other centers. Solutions based on drawing layer analysis, adopted to improve automation, frequently fail due to inconsistencies in drawing layer standards across different centers, often requiring redrawing the drawings, resulting in digital site reconstructions that are neither accurate nor universal. On the other hand, even with a site map, multi-source heterogeneous data such as equipment status, vehicle trajectories, package flow, video, and personnel data remain scattered across independent systems with delayed updates. It's difficult to aggregate data by on-site equipment and map it to specific devices, making it impossible to trace packages outside loading / unloading areas in core flow areas, requiring manual inspections to detect line congestion, and lacking quantitative data support for end-of-line scheduling. The physical layout and operational data cannot be linked on the same canvas, ultimately resulting in a visible layout but unclear operations, visible data but inability to locate specific equipment. Summary of the Invention

[0005] In order to enable the physical layout of the transfer center and multi-source operational data to be linked on a unified canvas, thereby improving the accuracy and versatility of digital site reconstruction and the real-time nature of on-site control, this application provides a digital twin visualization control method and system for logistics transfer centers.

[0006] Firstly, this application provides a digital twin visualization control method for logistics transfer centers, which adopts the following technical solution:

[0007] A digital twin visualization control method for logistics transfer centers includes the following steps:

[0008] S1. Receive the drawing data file converted from the on-site drawings at the transfer center. Using the blocks in the drawing data file as the parsing reference, traverse the block reference entities in the drawing data file. Based on the insertion point information of each block reference entity and the corresponding block definition, restore the physical coordinates and equipment type of each on-site device. Recursively traverse the dynamic blocks with visibility status and retain only the primitives corresponding to the visibility status. Based on the coordinate adjacency relationship of each on-site device restored, construct the equipment topology relationship between each on-site device and generate a digital canvas that carries each on-site device.

[0009] S2. Bind the corresponding field device identifier and service group to the graphic elements in the digital canvas, access multi-source heterogeneous data, aggregate the multi-source heterogeneous data according to the bound field device identifier and overlay it onto the corresponding graphic elements in the digital canvas, and update the overlay result in real time.

[0010] S3. In response to a business request for the digital canvas, determine the target graphic element and target device associated with the business request based on the binding relationship of graphic elements and the device topology relationship, and perform a visualization response corresponding to the business request on the target graphic element on the digital canvas.

[0011] By adopting the above technical solution, the blocks in the drawing data file are used as the analytical benchmark to restore the field equipment. This makes the construction of digital sites no longer dependent on manual drawing or the uniformity of layer standards. With the recursive traversal of dynamic blocks and the construction of equipment topology based on coordinate adjacency, the equipment layout consistent with the field can be automatically restored and adapted to drawings from different centers. On this basis, field equipment identifiers are bound to the graphic elements and multi-source heterogeneous data is aggregated and superimposed by equipment. This allows the operational data scattered in various systems to be located on specific equipment graphic elements. Then, when responding to business requests, the target equipment is located based on the graphic element binding and equipment topology and a visual response is executed. This allows the field physical layout and operational data to be linked on the same canvas, improving the accuracy and versatility of site restoration and the real-time nature of field control.

[0012] Optionally, the business request includes a package path restoration request. The visual response to the package path restoration request in S3 includes sub-steps S31-S34:

[0013] S31. Obtain the sorting device code and flow information of the target package based on the package path restoration request;

[0014] S32. Reconstruct the forward path starting from the sorting equipment code to obtain the forward path;

[0015] S33. Starting from the flow direction information, perform reverse path reconstruction to obtain the reverse path;

[0016] S34. Merge the forward and reverse paths to obtain the full-link path within the target package, and highlight the target primitives corresponding to the full-link path within the field on the digital canvas.

[0017] By adopting the above technical solution, the bidirectional restoration of sorting equipment coding in the forward direction and flow information in the reverse direction complements each other, enabling the reconstruction of the path of the package in the core circulation area outside the loading and unloading port, thus supplementing the on-site full-chain traceability capability.

[0018] Optionally, S32 includes sub-steps S321-S323:

[0019] S321. Locate the bound graphic element through the sorting equipment code and determine the line group to which the graphic element belongs;

[0020] S322. Starting from the line group, calculate the equipment chain through which the target package passes along the equipment topology;

[0021] S323. Find the nearest cameras to the devices in the device chain and mark the priority of each camera.

[0022] By adopting the above technical solution, the device chain is calculated step by step along the device topology and the nearest camera is matched, so that the forward path not only restores the device chain but also associates it with the video points that can be reviewed, thereby improving the verifiability of the tracing results.

[0023] Optionally, S33 includes sub-steps S331-S332:

[0024] S331. Locate the bound loading and unloading equipment elements using flow direction information;

[0025] S332. Starting from the loading and unloading equipment elements, reverse the calculation of the upstream equipment chain through which the target package passes along the equipment topology.

[0026] By adopting the above technical solution, the upstream equipment chain is reversed using the flow direction as the anchor point, which enables the completion of the package path in the outbound scenario where there is a lack of sorting equipment coding, thus complementing the forward reconstruction.

[0027] Optionally, the recursive traversal of the dynamic block containing visibility states in S1 includes: traversing all visibility states in the block definition of the dynamic block, processing only the primitives corresponding to the visibility states, and recursively performing block parsing on the nested sub-blocks composed of other blocks in the dynamic block.

[0028] By adopting the above technical solution, the multiple visibility states of dynamic blocks and the recursive processing of nested sub-blocks enable complex combined devices to be completely and accurately restored, thereby improving the coverage of the parsing.

[0029] Optionally, S1 also includes: calculating the physical coordinates of each field device based on the insertion point information, scaling ratio and unit conversion factor of each referenced entity; calibrating the physical coordinates of each field device with the preset device entrance as the coordinate origin; and dividing each field device into its corresponding functional area according to the calibrated coordinate range.

[0030] By adopting the above technical solution, through unit conversion, origin calibration and coordinate range partitioning, the restoration results have a unified coordinate reference and functional area affiliation, which facilitates subsequent device binding and regional display.

[0031] Optionally, S2 adopts a two-layer caching strategy that combines bypass caching with timed preheating for accessing multi-source heterogeneous data. This includes: responding to data query requests by directly returning cached data when the cache is hit, and concurrently requesting multiple external data sources and writing the aggregation results to the cache when the cache is not hit; and periodically traversing the binding relationship of graph elements through a timed task to concurrently request multiple external data sources and write the aggregation results to the cache in batches.

[0032] By adopting the above technical solution, combining on-demand querying with timed preheating, frequently accessed data can be cached while cold data is prepared in advance through preheating, reducing access latency for multi-source data.

[0033] Optionally, the multi-source heterogeneous data includes the motor frequency of the production line equipment. S2 also includes: converting the motor frequency into the actual operating speed of the production line equipment based on the speed conversion coefficient corresponding to the production line equipment.

[0034] By adopting the above technical solution, the collected motor frequency is converted into the actual speed of the production line according to the equipment coefficient, so that the operating status of the production line can be quantitatively characterized, providing a basis for judging speed anomalies and blockages.

[0035] Optionally, the business request also includes a line operation status monitoring request. The visualization response in S3 for the line operation status monitoring request includes: in response to the ratio of the actual operating speed of the line equipment to the standard speed being lower than a preset speed threshold, marking the target element corresponding to the line equipment as a speed abnormality state; in response to a blockage event generated by meeting preset blockage judgment conditions, marking the target element corresponding to the line equipment as a blockage state with higher priority than the speed abnormality state, determining the location of the blockage event and nearby cameras, and concurrently pushing blockage warnings through the voice channel and message channel.

[0036] By adopting the above technical solution, speed anomalies and blockages are prioritized and marked, and then associated with the nearest camera. Combined with dual-channel push notifications of voice and messages, line blockages can be automatically detected and promptly addressed by personnel.

[0037] Optionally, the business request may also include a clearance inventory risk assessment request. The visual response to the clearance inventory risk assessment request in S3 includes sub-steps S35-S38:

[0038] S35. Determine the amount of data to be processed through aggregated queries of package data;

[0039] S36. Calculate the capacity gap by combining vehicle scheduling data with historical loading data;

[0040] S37. Determine the risk level of leftover cargo based on the capacity gap, and mark the target elements corresponding to the flow directions with leftover cargo risk as abnormal on the digital canvas;

[0041] S38. Receive flow direction adjustment operations for tail cargo risks via mobile terminal, and update the scheduling status after the flow direction adjustment operation is confirmed.

[0042] By adopting the above technical solution, the risk level of the remaining cargo can be calculated from the amount to be operated and the capacity gap and marked on the canvas. Then, the flow direction can be adjusted through the mobile terminal, so that the scheduling of the remaining cargo can obtain quantitative data support and an executable closed loop.

[0043] Optionally, the priority marking of graphic elements includes marking the nearest graphic element as the first priority and highlighting it with a first display style, marking the remaining graphic elements as the second priority and highlighting them with a second display style, and rendering graphic elements that do not hit the path with a default background style.

[0044] By adopting the above technical solution, the core path and related paths are distinguished by a hierarchical display style, making the tracing results clear at a glance on the canvas.

[0045] Optionally, when S34 merges forward and reverse paths, it processes cross-floor paths into floor segments and determines whether each segment is in the same location based on the continuity of the line.

[0046] By adopting the above technical solution, the cross-floor path is segmented, so that the wrapping path of a multi-floor site can still be continuously restored.

[0047] Optionally, packet congestion alerts will be downgraded to being pushed only via the messaging channel when the voice channel push fails.

[0048] By adopting the above technical solutions, the early warning channel has degradation tolerance, so that the early warning can still be delivered when a single channel is abnormal.

[0049] Optionally, the two-layer caching strategy caches null values ​​for query results of non-existent device identifiers and sets different lifespans for different categories of cached data.

[0050] By adopting the above technical solutions, null value caching and tiered lifespan reduce the impact of invalid queries on external systems.

[0051] Optionally, S1 can revert to the previous historical snapshot of the digital canvas when the drawing data file is corrupted, and the digital canvas can be published from draft state to published state and written to the historical snapshot after fine editing.

[0052] By adopting the above technical solutions, the version snapshot and rollback mechanism ensures the recoverability of the digital canvas in case of anomalies.

[0053] Optionally, if the risk assessment of leftover goods is not confirmed within a preset time period, the risk level will be automatically upgraded and a higher-level manager will be notified, and duplicate scheduling for the same flow direction will be merged.

[0054] By adopting the above technical solution, unconfirmed scheduling is automatically upgraded and duplicate scheduling is merged, thus avoiding the omission of risks or the occurrence of scheduling conflicts.

[0055] Optionally, the drawing data file can be converted from a DWG format drawing or a vector drawing file.

[0056] By adopting the above technical solution, it is compatible with multiple drawing source formats, thus expanding the applicability of digital site reconstruction.

[0057] Optionally, path reconstruction also combines the mapping relationship between the package RFID tag reading records and device coordinates to locate the package location.

[0058] By adopting the above technical solution, radio frequency reading records are introduced as a supplementary data source to improve the accuracy of path reconstruction.

[0059] Optionally, the criteria for determining blockage may also include the analysis results of the motor's current harmonic components or the line vibration data.

[0060] By adopting the above technical solutions, current harmonics or vibration dimensions are introduced to enrich the basis for determining the blockage.

[0061] Optionally, the risk level of the remaining cargo can be determined by a classification model trained on historical scheduling data.

[0062] By adopting the above technical solution, the trained classification model can replace rule-based judgment, thereby improving the adaptability of risk assessment.

[0063] Optionally, the visual response of the digital canvas can be presented through web graphics rendering or a 3D digital twin scene built with a 3D engine.

[0064] By adopting the above technical solution, both two-dimensional rendering and three-dimensional scene presentation methods are supported, adapting to different display needs.

[0065] Secondly, the digital twin visualization control system for logistics transfer centers provided in this application adopts the following technical solution:

[0066] A digital twin visualization control system for a logistics transfer center includes:

[0067] The digital canvas restoration module is configured to receive drawing data files converted from on-site drawings at the transfer center, restore the physical coordinates and equipment types of each on-site device using the blocks in the drawing data files as the parsing reference, recursively traverse dynamic blocks with visibility states and retain only the primitives corresponding to the visibility states, construct the equipment topology relationship between each on-site device, and generate a digital canvas that carries each on-site device.

[0068] The data fusion module is configured to bind the graphic elements in the digital canvas to the corresponding field equipment identifiers and service groups, access multi-source heterogeneous data, aggregate the multi-source heterogeneous data according to the bound field equipment identifiers, and overlay them onto the corresponding graphic elements in the digital canvas.

[0069] The business response module is configured to respond to business requests for the digital canvas by determining the target graphic elements and target devices associated with the business request based on the binding relationship between graphic elements and the device topology relationship, and performing a visual response on the target graphic elements on the digital canvas corresponding to the business request.

[0070] In summary, this application includes at least one of the following beneficial technical effects:

[0071] 1. By using blocks in the drawing data file as the parsing benchmark and recursively traversing dynamic blocks, and automatically constructing equipment topology relationships by combining coordinate adjacency, the digital site can automatically restore the equipment layout consistent with the site without relying on manual drawing and layer standards, thereby improving the accuracy of site restoration and cross-center versatility.

[0072] 2. By binding field equipment identifiers to canvas elements and aggregating and overlaying multi-source heterogeneous data by device, combined with a two-layer strategy of bypass caching and timed preheating, the operational data scattered across various systems is placed on specific equipment elements and updated in real time, realizing the linkage between field physical layout and operational data on a unified canvas.

[0073] 3. By responding to business requests based on primitive binding and device topology, the entire parcel network is reconstructed through forward and reverse bidirectional path restoration; hierarchical marking and multi-channel push enable automatic discovery and reach of parcel blockages on the line; and capacity gap estimation and mobile terminal flow adjustment support the scheduling of tail cargo. This allows on-site management to obtain quantitative data support and an executable closed loop in the traceability, early warning and scheduling stages. Attached Figure Description

[0074] Figure 1 A flowchart illustrating the digital twin visualization control method for logistics transfer centers provided in this application embodiment.

[0075] Figure 2 This is a schematic diagram of the digital canvas restoration process provided in an embodiment of this application.

[0076] Figure 3 This is a schematic diagram of the package path restoration process provided in an embodiment of this application.

[0077] Figure 4 This is a schematic diagram of the architecture of the digital twin visualization control system for logistics transfer centers provided in the embodiments of this application. Detailed Implementation

[0078] The present application will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the scope of the application.

[0079] This application discloses a digital twin visualization and control method for logistics transfer centers. This method automatically reconstructs a digital station using blocks in the site drawings as the analytical benchmark, binds site equipment to canvas elements and overlays multi-source operational data, and then responds to business requests to the canvas based on the binding relationships and equipment topology, achieving linkage between the site's physical layout and operational data on a unified canvas. First, the coordinates, types, and topology of each site device are reconstructed from blocks in the drawing data file to generate a digital canvas. Then, site equipment identifiers are bound to canvas elements, and multi-source heterogeneous data is overlaid onto the corresponding elements. Finally, corresponding visualization responses are executed in response to business requests to the digital canvas. Thus, the site's physical layout and real-time operational data can be presented in association on a unified canvas, supporting operational control such as parcel path tracking, line congestion warnings, and risk scheduling of leftover goods.

[0080] To facilitate understanding, let's first illustrate the limitations of existing practices using a transit center scenario. A transit center's matrix sorting area contains dozens of cross-belt sorters, racks, and telescopic conveyors. Operations personnel need to locate the path of a lost package within the facility. Currently, the layout diagram is manually drawn in a 2D drawing tool according to layer standards, while operational data is scattered across multiple independent system reports related to equipment, packages, and transportation. When drawings from different centers do not follow a unified layer naming convention, layer-based analysis struggles to reliably identify equipment. Furthermore, the layout diagram and operational data are disconnected, preventing operations personnel from tracing back which equipment a package passed through and which monitoring screens it corresponded to from a single view.

[0081] The following is a description of some of the terms used in the embodiments of this application.

[0082] The digital canvas is a visual carrier that carries the graphic elements and spatial layout of various field equipment in the transfer center. Each graphic element in the canvas corresponds to a field device and records the physical coordinates, device type and connection relationship of the device.

[0083] A block is a reusable graphic definition unit in a drawing data file. A single block definition can be inserted multiple times to represent the same type of device. Dynamic blocks with visibility states can carry multiple display modes within a single block definition, while nested sub-blocks are blocks composed of other blocks.

[0084] Equipment topology is the connection relationship between various field devices determined by coordinate adjacency and connecting line primitives, used to characterize the reachable flow path wrapped between devices.

[0085] The binding relationship is the correspondence established between the graphic elements in the digital canvas and the field equipment identifiers and service groups they represent, and multi-source heterogeneous data is superimposed onto the corresponding graphic elements accordingly.

[0086] The following section details this method from aspects such as the restoration of digital data stations, the binding and overlay of multi-source data, and the visual response to business requests. (Refer to...) Figure 1 This method includes steps S1 to S3.

[0087] S1. Receive the drawing data file converted from the on-site drawings at the transfer center. Using the blocks in the drawing data file as the parsing reference, traverse the block reference entities in the drawing data file. Based on the insertion point information of each block reference entity and the corresponding block definition, restore the physical coordinates and equipment type of each on-site device. Recursively traverse the dynamic blocks with visibility status and retain only the primitives corresponding to the visibility status. Based on the coordinate adjacency relationship of each on-site device obtained from the restoration, construct the equipment topology relationship between each on-site device and generate a digital canvas that carries each on-site device.

[0088] In the specific implementation, the drawing data file can adopt the DXF format, which consists of a header segment, a block segment, and an entity segment. The header segment contains system variables that determine the drawing units. The system variable INSUNITS takes values ​​of 1, 4, and 6, representing inches, millimeters, and meters, respectively, to determine the unit conversion factor. The block segment contains all block definitions, recording the reference point coordinates, contained elements, attribute definitions, and visibility status for each block definition. The entity segment contains all element entities, where the block insertion entity representing equipment calls the block resolution logic, while elements such as line segments, polylines, and arcs are treated as connection lines between equipment. For each block insertion entity in the block segment, its block name is read and the corresponding block definition is found. The insertion point coordinates, scaling ratio, and rotation angle are read, and attribute values ​​such as equipment number and height are extracted from the block's attribute definition. Then, the type of equipment is determined through the mapping from block name to equipment type. After reconstructing the coordinates of each device, the device pairs are traversed. When the center distance between two devices is less than a preset connection threshold and there is a connecting line element between them, a connection relationship is established between them, thereby constructing the device topology relationship. This connection threshold can be set to 5 meters. Then, all device coordinates are traversed to construct the canvas boundary and generate a digital canvas that carries each device. The reconstructed canvas content can be stored in a canvas information table. This table includes fields such as transfer center code, site code, unique canvas identifier, save status, and canvas element content recorded in JSON format, including the coordinates of each device, rotation angle, and scaling ratio. The save status can take values ​​such as draft, saved, and published.

[0089] For example, in the drawing data file of a certain transfer center, the system variable INSUNITS in the header has a value of 6 and a unit conversion factor of 1.0. The block section constructs an index of 156 block types, including 24 dynamic blocks with visibility status. The entity section identifies 1876 block insertion entities. The parsing takes about 18 seconds and generates canvas primitive content. The consistency between the restored equipment layout and the actual layout on site can reach about 99%.

[0090] Building upon the block parsing described above, the processing of dynamic blocks containing visibility states can be further refined. A dynamic block, within a single block definition, carries multiple display states through its visibility states. For example, a dynamic block for the same conveyor line can contain multiple visibility states such as straight ahead and turning. For such dynamic blocks, all visibility states recorded in its block definition are traversed, and only the primitives corresponding to states marked as true are processed, thus avoiding redundancy caused by restoring primitives with invisible states. When a dynamic block contains nested sub-blocks composed of other blocks, the aforementioned block insertion entity parsing and visibility filtering are recursively performed on the nested sub-blocks until the lowest-level basic primitives are expanded. As an optional implementation, visibility state filtering can also be based on a pre-configured whitelist of visibility state names; the recursive expansion of nested sub-blocks can be implemented either through recursive calls or by using an explicit stack structure in an iterative manner to adapt to situations with deep nesting levels.

[0091] The physical coordinates of each device are determined through coordinate transformation, origin calibration, and region division. Based on the insertion point coordinates, scaling factor, and unit conversion factor of the block insertion entity, the physical coordinates of each field device in the world coordinate system are calculated. For example, the abscissa of a device in the world coordinate system is calculated using the following formula:

[0092]

[0093] in, The x-coordinate of the device in the world coordinate system. The x-coordinate of the insertion point of the block-inserted entity. The unit conversion factor is determined by the drawing units. The horizontal scaling factor for the inserted entity is used, and the vertical coordinate is calculated in the same way from the vertical coordinate of the insertion point, the unit conversion factor, and the vertical scaling factor. After obtaining the coordinates of each device, the coordinates of all devices are calibrated using a preset device entrance as the coordinate origin, for example, using the location of the telescopic conveyor entrance as the coordinate origin. Then, each device is assigned to its corresponding functional area according to the calibrated horizontal coordinate range; for example, devices with horizontal coordinates less than 0 are assigned to the unloading area, devices with horizontal coordinates between 0 and 200 are assigned to the matrix area, and devices with horizontal coordinates greater than 200 are assigned to the loading area. The above digital canvas restoration process is as follows: Figure 2 As shown.

[0094] In some embodiments, abnormal compatibility handling is provided for drawing data files. When a drawing data file is corrupted, its header signature is verified, and if verification fails, the process reverts to the previous version snapshot saved in the canvas history snapshot table. A timeout threshold is set for drawing parsing, which can be 30 seconds. After the timeout, parsing is terminated and the administrator is notified. When the names of dynamic blocks such as conveyor lines and turning machines are parsed abnormally and cannot be identified by block name, the process reverts to compatibility logic based on layer information to identify the device. When the block definitions of devices such as telescopic cranes are non-standard and exhibit abnormal length, width, or coordinates, the block is mirrored or split into multiple blocks before parsing to adapt to non-standard blocks. After fine-tuning, the digital canvas can be published from a draft state to a published state. This fine-tuning includes correcting misaligned devices, deleting useless devices, etc., and the canvas content at the time of publication is written to the canvas history snapshot table, which records each canvas content snapshot according to the canvas identifier and change time.

[0095] In addition to being converted from DXF format drawings, in some embodiments, drawing data files can also be obtained directly from DWG format drawings using corresponding parsing tools, or from vector drawing files such as PDF by extracting vector paths, thus ensuring compatibility with multiple drawing source formats.

[0096] After generating a digital canvas that carries various field devices, bind the field devices to the canvas elements and overlay multi-source operational data.

[0097] S2. Bind the corresponding field device identifier and service group to the graphic elements in the digital canvas, access multi-source heterogeneous data, aggregate the multi-source heterogeneous data according to the bound field device identifier and overlay it onto the corresponding graphic elements in the digital canvas, and update the overlay result in real time.

[0098] In the specific implementation, binding relationships are established for each graphic element in the canvas, recording the graphic element identifier, graphic element type, corresponding field equipment identifier, and the business group to which it belongs. Graphic element types may include cross-belt sorters, side-sliding machines, telescopic machines, swing wheels, conveyor lines, cameras, etc. Business groups may include area groups such as matrix upper-level lines and unloading areas. The binding relationships can be stored in a graphic element binding table and a business group table. The graphic element binding table records fields such as graphic element identifier, graphic element type, and equipment identifier, while the business group table records the group identifier, group name, and a list of graphic element identifiers contained in that group. Based on the binding relationship, multi-source heterogeneous data is accessed. This accessed multi-source heterogeneous data may include: real-time equipment status data acquired via fieldbus protocol, such as motor frequency, current, and voltage, with a collection cycle of once every 10 seconds; vehicle trajectory data, including vehicle dispatch time, departure node, and expected destination; parcel operation data, including operation type, operation time, and operation organization, such as dispatch scan 210, loading scan 231, and arrival scan 520; personnel team data, including team number, number of personnel, team leader, and supervisor; and camera video stream data. The above multi-source heterogeneous data is aggregated according to the bound field equipment identifier and overlaid onto the corresponding graphic elements, with the overlay results updated in real time. The aggregated data can be written to a cache by graphic element or region, for example, cached using equipment status, line indicators, region efficiency, canvas data, and canvas summary as keys, with different lifespans set for different data categories.

[0099] Operational data overlaid on the canvas can be updated to the front end via a real-time push channel, which can be a full-duplex WebSocket channel. In other embodiments, the real-time push channel can also use a streaming interface that sends server events or remote procedure calls to push data from the server to the front end. The aggregated data can be written to an in-memory cache, such as a Redis-based remote cache. In other embodiments, Memcached or a local cache located within the service process can also be used to adapt to different deployment scales and access latency requirements.

[0100] For example, by using a scheduled task to traverse the primitive binding table every 5 minutes and concurrently request line equipment data, package operation data, and vehicle scheduling data, the aggregated data is written to the cache, ensuring that the update delay of the operation data overlaid on the canvas does not exceed 3 seconds.

[0101] To reduce access latency for multi-source data, a two-layer caching strategy combining bypass caching and scheduled preheating can be adopted when accessing heterogeneous multi-source data. This strategy includes two paths: on-demand querying and scheduled preheating. In the on-demand querying path, in response to a data query request, the cache is checked first. If the cache is hit, the cached data is returned directly. If the cache is not hit, multiple external data sources are requested concurrently, such as production line equipment data sources, package data sources, vehicle scheduling data sources, and personnel organization data sources. The returned data is processed and aggregated, written to the cache, and a lifespan is set before the query results are returned. In the scheduled preheating path, a scheduled task periodically traverses the binding relationships of graphic elements, concurrently requests multiple external data sources, and writes the aggregated results to the cache in batches. Thus, frequently accessed data can directly hit the cache, while cold data is prepared in advance through preheating.

[0102] The accessed multi-source heterogeneous data includes the motor frequencies of the production line equipment. These motor frequencies can be converted into the actual operating speed of the production line equipment based on the speed conversion coefficient corresponding to the equipment. For example, the actual operating speed of the production line equipment is calculated using the following formula:

[0103]

[0104] in, This refers to the actual operating speed of the production line equipment. The collected motor frequency, The speed conversion coefficient of this production line equipment is uniquely maintained according to the motor code. These are the conversion parameters from frequency to speed. For example, if the motor frequency of a certain production line is 50Hz, its speed conversion coefficient and conversion parameters result in a calculated actual operating speed of 1.0m / s. When the motor frequency decreases, the calculated actual operating speed decreases accordingly, thus enabling a quantitative characterization of the production line's operating status and providing a basis for subsequent speed anomaly and blockage detection.

[0105] To mitigate the impact of invalid queries on external systems, a two-tiered caching strategy can cache null values ​​for query results of non-existent device identifiers, and set different lifespans for different categories of cached data. For example, the cache lifespan for device status data can be set to 60 seconds, line indicator data to 30 seconds, and regional efficiency data and canvas summary data to 300 seconds; for non-existent device identifiers, query results are cached as null values ​​with a lifespan of 10 seconds. A data interruption alarm is triggered when data from the external data source is interrupted for more than 30 seconds, and the previously cached data is returned as a degradation when a query times out.

[0106] Based on the canvas primitives being bound to field devices and overlaid with multi-source data, respond to business requests for the digital canvas.

[0107] S3. In response to a business request for the digital canvas, determine the target graphic element and target device associated with the business request based on the binding relationship of graphic elements and the device topology relationship, and perform a visualization response corresponding to the business request on the target graphic element on the digital canvas.

[0108] In practical implementation, business requests may include parcel path restoration requests, production line operation status monitoring requests, and tail-end cargo risk assessment requests. When responding to business requests, the associated target graphic elements and target devices are determined based on the binding relationship between graphic elements and the device topology relationship. Corresponding visual responses are then executed for the target graphic elements, such as marking target graphic elements with different display states: gray when the production line device is not running, blue when the actual operating speed reaches the standard speed, yellow when the actual operating speed is lower than the standard speed but has not reached the congestion level, and red when congestion occurs (red has the highest priority). When multiple anomalies occur simultaneously, red is given priority. For graphic elements with anomalies, their anomaly information can be bound in JSON format and displayed as an anomaly identifier on the canvas. Jumping from the anomaly details to the corresponding anomaly graphic element location on the canvas is also supported.

[0109] In some embodiments, the visualization response to the digital canvas can be presented through Web graphics rendering or through a 3D digital twin scene built by a 3D engine. For example, 2D rendering can be performed on the front end based on WebGL, or a 3D digital twin scene can be built using a 3D engine to support immersive viewing of the operational status of the transfer center.

[0110] One type of business request is a parcel path restoration request. The flow of parcels within a transfer center typically includes two segments: forward and reverse. The forward flow starts from the unloading telescopic conveyor, passing through conveyor lines, cross-belt sorters or side slides, wheel pulleys, and chutes, to the loading telescopic conveyor; the reverse flow starts from the loading dock, passing through the loading telescopic conveyor and conveyor lines. One-way tracing alone cannot cover the complete flow of parcels within the center; therefore, bidirectional path restoration is performed on parcel path restoration requests and then merged to reconstruct the entire link path within the center. The visual response to the parcel path restoration request in S3 includes sub-steps S31 to S34.

[0111] S31. Obtain the sorting equipment code and flow information of the target package based on the package path restoration request. For example, after entering the package's tracking number, call the package information interface to obtain basic information such as the package's tracking number, weight, volume, flow direction, and sorting equipment code.

[0112] S32. Reconstruct the forward path starting from the sorting equipment code to obtain the forward path.

[0113] S33. Starting from the flow direction information, perform reverse path reconstruction to obtain the reverse path.

[0114] S34. Merge the forward and reverse paths to obtain the full-link path within the target package, and highlight the target primitives corresponding to the full-link path on the digital canvas. When merging the forward and reverse paths, for the lower-level lines of the matrix, bind them according to their respective regions and only display the lower-level lines of the matrix that are in the same region and the same site as the target package, so as to avoid irrelevant lines interfering with the path presentation.

[0115] For example, for a lost package, after entering its tracking number, the sorting equipment code DWS_SUZ_001 and its destination Shanghai are obtained. The forward path is reconstructed starting from sorting equipment code DWS_SUZ_001, and the reverse path is reconstructed starting from the destination Shanghai. The merged paths provide the package's complete on-site path, highlighting the corresponding elements. The average response time for a single query is approximately 2.8 seconds, and the accuracy of reconstructing the on-site flow trajectory can reach over 95%, thus reducing the location time of lost items to minutes. As an optional implementation method, in addition to equipment topology calculation, package path reconstruction can also be assisted by relay tracking from surveillance videos.

[0116] Specifically, forward path restoration in S32 includes sub-steps S321 to S323.

[0117] S321. Locate the bound graphic element using the sorting equipment code and determine the line group to which the graphic element belongs. For example, look up the bound graphic element identifier in the graphic element binding table using the sorting equipment code, and then determine the line group to which the graphic element belongs using the business grouping table.

[0118] S322. Starting from the line group, calculate the equipment chain through which the target package passes along the equipment topology. For example, starting from the line group, calculate the equipment chain through which the package passes along the connection relationship of each device in the equipment topology segment by segment according to the device adjacency distance and the connecting line elements.

[0119] S323. Find the nearest cameras to the devices in the device chain and assign priority to each camera. For example, find the three nearest cameras to the devices in the device chain, mark the nearest camera as the first priority, and the other two as the second priority.

[0120] Specifically, refer to Figure 3 The reverse path restoration in S33 includes sub-steps S331 to S332.

[0121] S331. Locate the bound loading / unloading equipment elements by flow direction information. For example, locate the bound telescopic crane or parking space elements by flow direction name, and the organization name can be used for matching during the search.

[0122] S332. Starting from the loading and unloading equipment element, calculate the upstream equipment chain through which the target package passes by in reverse along the equipment topology. For example, starting from the telescopic conveyor element, calculate the upstream equipment chain through which the package passes by segment by segment in reverse along the equipment topology, and similarly bind the nearest cameras to the equipment in the upstream equipment chain.

[0123] The graphic elements in the path reconstruction results can be highlighted with different display styles according to their priority. For example, the closest graphic element is marked as the first priority and highlighted with the first display style, and the corresponding camera is marked with a specific mark. The remaining graphic elements are marked as the second priority and highlighted with the second display style. Graphic elements that do not match the path are presented with the default background style. The first display style, the second display style, and the default background style can be yellow, blue, and gray, respectively, and the specific mark for the camera can be green.

[0124] In some embodiments, for cameras marked in the path reconstruction results, corresponding video playback links are further generated, and the name of the camera and the video playback entry are displayed in a floating manner on the canvas. The playback end time of the video playback link can be automatically calculated according to the configured duration of the area to which the camera belongs, thereby supporting direct viewing of the monitoring screen of the passing devices on the reconstructed path.

[0125] In some embodiments, the user's query history for package path restoration requests is retained, and the most recent query history is displayed on the page, such as the most recent 3 query history, so as to allow for repeated backtracking.

[0126] In other embodiments, when merging forward and reverse paths, the path spanning multiple floors is segmented by floor. For example, the path from the first floor through the second floor and back to the first floor is divided into three segments for processing, and the continuity of the line is used to determine whether each segment is in the same location.

[0127] In addition to device topology deduction, as an optional implementation method, route reconstruction can also introduce the reading records of package RFID tags as a supplementary data source. The package location can be directly located through the mapping relationship between RFID reading points and device coordinates to improve the accuracy of route reconstruction.

[0128] In addition to package path reconstruction requests, business requests also include line operation status monitoring requests. The visual response to these requests in S3 includes speed anomaly flags and package blockage warnings. The speed status is determined based on the ratio of the actual operating speed of the line equipment to the standard speed. For example, the speed ratio is calculated using the following formula:

[0129]

[0130] in, The ratio of speeds, This refers to the actual operating speed of the production line equipment. This is the standard speed of the production line. In response to a speed ratio lower than a preset speed threshold, the target element corresponding to the production line is marked as having an abnormal speed state. This speed threshold can be 0.85, and the abnormal speed state can be marked in yellow. In some embodiments, the determination of the speed state can be based not only on the speed ratio but also on the ratio of real-time current to reference current for auxiliary determination. The current ratio is calculated according to... Calculation, where This is the current ratio. For real-time current, The reference current is used. Furthermore, in response to a blockage event generated based on preset blockage judgment conditions, the target graphic element corresponding to the line equipment is marked as a blockage state with higher priority than the speed anomaly state. The blockage state can be marked in red. The location of the blockage event and nearby cameras are determined, and a blockage warning is pushed concurrently through the voice and message channels. A blockage event can be generated by real-time judgment of the line equipment status data, for example, represented by a specific blockage event identifier. The generated blockage event includes the line identifier, location, nearby cameras, and a warning message. Nearby cameras can be determined based on the nearest cameras calculated from the graphic element coordinate distance. The voice channel can call the voice warning service and query the warning contact person by transfer center code and port entry / exit type before making a call to broadcast the warning content. The message channel can send a group message and remind the supervisor. When assembling a blockage warning, recommended actions can be further determined based on the type of blockage event, and the line location, nearby cameras, and recommended actions are written into the warning message for quick response by the receiving personnel.

[0131] For example, in a monitoring scenario, at the 5th second, the motor frequency of a certain production line device drops to 42Hz, translating to an actual operating speed of approximately 0.84m / s. This speed ratio (ratio to the standard speed of 1.0m / s) is approximately 0.84, and the current increases from approximately 10.2A to 12.8A, resulting in a current ratio of approximately 1.25. At the 10th second, a bag blockage event is generated. At the 12th second, the location coordinates of this event are determined to be approximately x-coordinate 85.3 and y-coordinate 42.1, and the three nearest cameras are identified, for example, as CAM_12, ​​CAM_13, and CAM_14. At the 13th second, an alert is simultaneously pushed via voice and message channels. The entire process takes approximately 170 seconds. Therefore, the alert response time for bag blockages can be reduced to no more than 10 seconds, and the bag blockage handling time is shortened from approximately 15 minutes to less than 5 minutes, resulting in a corresponding increase in sorting line efficiency of approximately 30%.

[0132] In the push notification of packet blocking alerts, if the voice channel push fails, it can be downgraded to push only through the message channel to ensure that the alert can still reach people when a single channel is abnormal; if the alert contact person is not configured, the default supervisor contact method can be used.

[0133] In some embodiments, in addition to the speed ratio and current ratio, the blockage determination criteria may also include the analysis results of the motor's current harmonic components or the vibration data of the line, such as detecting abnormal frequency components by performing frequency domain transformation on the current, or collecting vibration sensor data of the line, to enrich the basis for blockage determination.

[0134] In addition to the above requests, business requests also include requests for risk assessment of clearance inventory. The visual response to the risk assessment of clearance inventory request in S3 includes sub-steps S35 to S38.

[0135] S35. Determine the amount to be processed through aggregated queries of parcel data. For example, the amount to be processed includes the parcel volume collected at the branch and the parcel volume transferred at the center. The parcel volume collected at the branch is the number of parcels collected within a certain period of time that have not been dispatched from the first center and whose destination center is not this center. The parcel volume transferred at the center is the number of parcels that are expected or actually arrive before the end of the current or next shift and have not been scanned for dispatch at this center. In some embodiments, the aggregated query of parcel data also includes statistically analyzing the distribution of dispatch volume from the destination to the next station, that is, grouping historical flow records by destination and next node to statistically analyze the dispatch volume of each destination to the next station, in order to help determine the flow of leftover goods.

[0136] S36. Calculate the capacity gap by combining vehicle scheduling data and historical loading data. For example, the capacity gap is calculated using the following formula:

[0137]

[0138] in, Due to the capacity gap, The quantity to be operated. For the scheduled loading volume, The available load capacity is determined based on the available vehicles and the historical maximum load capacity of vehicles.

[0139] S37. Determine the risk level of leftover cargo based on the capacity gap, and mark the target elements corresponding to the flow directions with leftover cargo risk as anomalies on the digital canvas. For example, when the capacity gap is greater than 0, it is determined that there is leftover cargo risk and the corresponding risk level is marked.

[0140] S38. Receive flow direction adjustment operations for tail cargo risks via mobile terminal, and update the scheduling status after the flow direction adjustment operation is confirmed.

[0141] In some embodiments, the risk assessment of leftover cargo is performed on two timed paths: pre-assessment and in-process monitoring. Pre-assessment is triggered at a longer interval, such as once every 30 minutes, to query the amount of cargo to be operated, vehicle information and historical maximum loading capacity and calculate the capacity gap to predict risks. In-process monitoring is triggered at a shorter interval, such as once every 5 minutes, to query the real-time volume of goods and vehicle data and update the flow-level anomaly flag accordingly.

[0142] For example, if a certain route has a collection volume of 5000 pieces at a network point and a central transfer volume of 3000 pieces, then the pending processing volume is 8000 pieces; the already scheduled loading volume is 4000 pieces, and the available loading volume is 2000 pieces. Substituting these values ​​into the above formula, the capacity gap is:

[0143]

[0144] The capacity gap is 2,000 pieces; the three vehicles on the backup route have a total volume of approximately 180 m³ and a historical maximum loading capacity of approximately 3,000 pieces. Since the capacity gap is greater than zero, this flow direction is deemed to have a high risk of leftover cargo and is marked as abnormal on the canvas. Relevant personnel confirm the flow adjustment plan via mobile device and update the scheduling status. As a result, the secondary transshipment delay rate of leftover cargo can be reduced by approximately 80%, departure time can be improved by approximately 40%, and the decision-making and execution cycle can be shortened by approximately 50%.

[0145] In other embodiments, when a scheduling plan is not confirmed within a preset time period, the risk level is automatically upgraded and a higher-level manager is notified. The preset time period can be 30 minutes. When there are duplicate scheduling for the same flow direction, the duplicate scheduling plans are merged to avoid risks being missed or scheduling conflicts occurring.

[0146] In addition to determining the risk level of leftover cargo based on capacity gaps, as an optional approach, it can also be determined by a classification model trained on historical scheduling data. For example, a classification model such as a gradient boosting tree can be trained using historical scheduling data as the training set. The classification model can automatically learn the risk characteristics of each flow direction to replace rule-based risk assessment.

[0147] The aforementioned digital twin visualization management and control method for logistics transfer centers can be implemented by a corresponding system. This application also provides a digital twin visualization management and control system for logistics transfer centers, which includes a digital canvas restoration module, a data fusion module, and a business response module.

[0148] The digital canvas restoration module is configured to receive drawing data files converted from on-site drawings at the transfer center, restore the physical coordinates and equipment types of each on-site device using the blocks in the drawing data files as the parsing benchmark, recursively traverse dynamic blocks with visibility states and retain only the primitives corresponding to the visibility states, construct the equipment topology relationship between each on-site device, and generate a digital canvas that carries each on-site device.

[0149] The data fusion module is configured to bind the graphic elements in the digital canvas to the corresponding field device identifiers and service groups, access multi-source heterogeneous data, aggregate the multi-source heterogeneous data according to the bound field device identifiers, and overlay it onto the corresponding graphic elements in the digital canvas.

[0150] The business response module is configured to respond to business requests for the digital canvas by determining the target graphic elements and target devices associated with the business request based on the binding relationship between graphic elements and the device topology relationship, and performing a visual response on the target graphic elements on the digital canvas corresponding to the business request.

[0151] In the specific implementation, refer to Figure 4 The aforementioned system can be implemented using a layered architecture, including, for example, a data source layer, a data access layer, a data processing layer, a data storage layer, a service layer, and a presentation layer. The data source layer provides line sensor data, vehicle trajectory data, video stream data, and personnel information; the data access layer accesses the aforementioned data through subscription, polling, retrieval, and database synchronization; the data processing layer includes various processing units responsible for the aforementioned restoration, fusion, path restoration, packet congestion warning, and risk assessment; the data storage layer stores canvas information, element bindings, business groupings, canvas historical version snapshots, and warning contact configurations, along with cached and time-series data; the service layer provides services externally through push notifications, APIs, video stream proxies, and notifications; and the presentation layer presents visual results through front-end rendering, operational dashboards, and mobile devices. All the above layers can communicate with each other via remote calls and secure transmission protocols.

[0152] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0153] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A digital twin visualization control method for logistics transfer centers, characterized in that, Includes the following steps: S1. Receive the drawing data file obtained by converting the on-site drawings at the transfer center. Using the blocks in the drawing data file as the parsing reference, traverse the block reference entities in the drawing data file. Based on the insertion point information of each block reference entity and the corresponding block definition, restore the physical coordinates and equipment type of each on-site device. Recursively traverse the dynamic blocks with visibility status and retain only the primitives corresponding to the visibility status. Based on the coordinate adjacency relationship of each on-site device obtained by restoration, construct the equipment topology relationship between each on-site device and generate a digital canvas that carries each on-site device. S2. Bind the corresponding field device identifier and service group to the graphic elements in the digital canvas, access multi-source heterogeneous data, aggregate the multi-source heterogeneous data according to the bound field device identifier and overlay it onto the corresponding graphic elements in the digital canvas, and update the overlay result in real time; S3. In response to a service request for the digital canvas, determine the target graphic element and target device associated with the service request based on the binding relationship of the graphic element and the device topology relationship, and perform a visualization response corresponding to the service request on the target graphic element on the digital canvas.

2. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, The business request includes a package path restoration request, and the visual response to the package path restoration request in S3 includes the following sub-steps: S31. Obtain the sorting device code and flow information of the target package according to the package path restoration request; S32. Reconstruct the forward path starting from the sorting equipment code to obtain the forward path; S33. Using the flow direction information as the starting point, perform reverse path reconstruction to obtain the reverse path; S34. Merge the forward path and the reverse path to obtain the full-link path within the target package, and highlight the target primitives corresponding to the full-link path within the field on the digital canvas.

3. The digital twin visualization control method for logistics transfer centers according to claim 2, characterized in that, S32 includes the following sub-steps: S321. The bound graphic element is located by the sorting device code, and the line group to which the graphic element belongs is determined; S322. Starting from the line group, calculate the equipment chain through which the target package passes along the equipment topology; S323. Find the nearest cameras to the devices in the device chain and mark the priority of each camera.

4. The digital twin visualization control method for logistics transfer centers according to claim 2, characterized in that, S33 includes the following sub-steps: S331. Locate the bound loading and unloading equipment graphic elements using the flow direction information; S332. Starting from the loading and unloading equipment elements, reverse the calculation of the upstream equipment chain through which the target package passes along the equipment topology.

5. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, The recursive traversal of the dynamic block containing a visibility state in S1 includes: traversing all visibility states in the block definition of the dynamic block, processing only the primitives corresponding to the visibility states, and recursively performing block parsing on the nested sub-blocks composed of other blocks in the dynamic block.

6. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, S1 further includes: calculating the physical coordinates of each field device based on the insertion point information, scaling ratio and unit conversion factor of each block reference entity; calibrating the physical coordinates of each field device with the preset device entrance as the coordinate origin; and dividing each field device into its corresponding functional area according to the calibrated coordinate range.

7. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, The S2 uses a two-layer caching strategy that combines bypass caching with timed preheating to access multi-source heterogeneous data. This includes: responding to data query requests by directly returning cached data when the cache is hit, and concurrently requesting multiple external data sources and writing the aggregation results to the cache when the cache is not hit; and periodically traversing the binding relationships of the graphic elements through a timed task, concurrently requesting the multiple external data sources and writing the aggregation results to the cache in batches.

8. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, The multi-source heterogeneous data includes the motor frequency of the production line equipment. S2 further includes: converting the motor frequency into the actual operating speed of the production line equipment based on the speed conversion coefficient corresponding to the production line equipment.

9. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, The service request also includes a line operation status monitoring request. The visualization response in S3 to the line operation status monitoring request includes: in response to the ratio of the actual operating speed of the line equipment to the standard speed being lower than a preset speed threshold, marking the target element corresponding to the line equipment as a speed abnormality state; in response to a packet blocking event generated by satisfying a preset packet blocking judgment condition, marking the target element corresponding to the line equipment as a packet blocking state with a higher priority than the speed abnormality state, determining the location of the packet blocking event and nearby cameras, and concurrently pushing packet blocking warnings through the voice channel and message channel.

10. The digital twin visualization control method for logistics transfer centers according to claim 1, characterized in that, The business request also includes a clearance risk assessment request, and the visual response to the clearance risk assessment request in S3 includes the following sub-steps: S35. Determine the amount of data to be processed through aggregated queries of package data; S36. Calculate the capacity gap by combining vehicle scheduling data with historical loading data; S37. Determine the risk level of the remaining cargo based on the capacity gap, and mark the target elements corresponding to the flow direction with remaining cargo risk as abnormal on the digital canvas; S38. Receive the flow direction adjustment operation for the risk of the leftover goods through the mobile terminal, and update the scheduling status after the flow direction adjustment operation is confirmed.

Images