Article loading method and device

Abstract

The invention discloses an article loading method and device, and relates to the technical field of logistics management. A specific embodiment of the method comprises the following steps: constructing a tree-shaped data structure of order data to be processed based on a three-dimensional index formed by an order starting point, a target point and a loading time window, and storing the order data to be processed in leaf nodes in the tree-shaped data structure respectively, and querying data in the leaf nodes according to target order parameters of the to-be-processed order data to obtain query results corresponding to the target parameters, generating order groups, and generating a loading scheme according to the loading constraint condition and the order groups. According to the embodiment, the articles can be accurately grouped, and the article loading scheme is quickly generated, so that the loading efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of logistics management technology, and in particular to a method and apparatus for loading goods. Background Technology

[0002] In the current maritime shipping landscape, less-than-container-load (LCL) shipping is a crucial operational activity. When carriers accept shipments of smaller quantities than a full container load from shippers, they categorize and organize the goods according to their nature and destination, then consolidate them into a single container for transport. LCL operations are primarily conducted manually, including screening orders with the same port of loading, port of destination, and time window, and combining and arranging goods based on compatibility factors (such as hazard class and temperature requirements). This process is relatively inefficient and generally lacks accuracy. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for loading goods, which can automatically group the goods in an order according to the parameters of the order and generate an accurate loading plan.

[0004] To achieve the above objectives, according to one aspect of the present invention, a method for loading articles is provided, comprising: A tree-like data structure for the order data to be processed is constructed based on a three-dimensional index of the order origin, destination, and loading time window. The order data to be processed is stored in the leaf nodes of the tree-like data structure, and the root node stores the order data to be processed for the current batch. Based on the target order parameters corresponding to the pending order data, query the pending order data in the leaf node, and generate order groups based on the query results corresponding to each target parameter; A loading scheme for orders to be processed is generated based on loading constraints and the order grouping.

[0005] Optionally, before constructing the tree data structure of the order data to be processed based on the three-dimensional index of the order origin, destination, and loading time window, the method further includes: Enhanced feature data of the items in the pending orders are generated based on the raw data of the pending orders; The pending order data is generated based on the enhanced feature data and the original data of the pending orders.

[0006] Optionally, the loading scheme for generating pending orders based on loading constraints and order grouping includes: For each order group, the orders in the order group are further grouped based on the attribute data of the items in the order group and the item compatibility rules to obtain the order subgroups in each order group; A loading scheme for the orders to be processed is generated based on the order grouping, the order subgrouping, and the loading constraints.

[0007] Optionally, the step of generating a loading scheme for orders to be processed based on the order group, the order subgroup, and the loading constraints includes: Based on the order subgroups and the loading constraints, a multi-objective genetic solution based on cargo rotation and stacking constraints is performed to obtain the loading scheme for each order group; The loading scheme for the orders to be processed is determined based on the loading scheme of each order group.

[0008] Optionally, before determining the loading scheme for the order to be processed based on the loading scheme of each order group, the method further includes: Collision detection is performed on the loading scheme of each order group based on the size data of the items in each order subgroup; The loading scheme for each order group is adjusted based on the collision detection results.

[0009] Optionally, the step of performing a multi-objective genetic solution based on cargo rotation and stacking constraints, based on the order subgroups and the loading constraints, to obtain the loading scheme for each order group, includes: The loading space is divided into layers to obtain multi-layer loading space; Based on the weight and size data of the items in the multi-layer loading space and the order subgroup, the items in the order subgroup are loaded in layers to obtain the items loaded in each layer of loading space; For each loading space, a multi-objective genetic solution based on cargo rotation and stacking constraints is performed based on the items loaded in each loading space and the loading constraints to obtain the loading scheme for each order group.

[0010] Optionally, generating order groups based on the query results corresponding to each target parameter includes: Based on the query results corresponding to each target parameter, determine at least one target parameter for each order to be processed; In response to the fact that there are multiple target parameters corresponding to any pending order, the first parameter corresponding to any pending order is determined based on the timeliness information of the pending order and the corresponding user information; In response to the fact that the number of target parameters corresponding to any pending order is one, the target parameter corresponding to any pending order is determined as the first parameter; Order groups are generated based on the first parameter corresponding to each pending order.

[0011] Optionally, the loading scheme for the orders to be processed includes multiple schemes, and the method further includes: The multiple order loading schemes are displayed in a three-dimensional way, allowing users to select the target order loading scheme based on these schemes.

[0012] According to another aspect of the present invention, an article loading device is provided, comprising: The construction module is used to construct a tree data structure of the order data to be processed based on the three-dimensional index of the order starting point, target point and loading time window. The order data to be processed is stored in the leaf nodes of the tree data structure, and the root node stores the order data to be processed in the current batch. The query module is used to query the pending order data in the leaf node according to the target order parameters corresponding to the pending order data, and generate order groups according to the query results corresponding to each target parameter; The generation module is used to generate a loading scheme for orders to be processed based on loading constraints and the order grouping.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the article loading method provided in the embodiments of the present invention.

[0014] According to another aspect of the present invention, a computer-readable medium is provided having a computer program stored thereon, which, when executed by a processor, implements the article loading method provided in the embodiments of the present invention.

[0015] According to another aspect of the present invention, a computer program product is provided, including a computer program that, when executed by a processor, implements the article loading method provided in the embodiments of the present invention.

[0016] One embodiment of the above invention has the following advantages or beneficial effects: it can construct a tree data structure based on the three-dimensional index determined by the acquired order data to be processed, and realize fast index query through the tree data structure to obtain order groups. This reduces the time complexity of the query operation from the traditional O(n²) to O(nlogn), significantly improving query speed and obtaining accurate query results. The order groups generated based on the query results, combined with preset loading constraints, help generate accurate loading schemes for the orders to be processed, greatly improving the efficiency of loading scheme generation.

[0017] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0018] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein: Figure 1 This is a schematic diagram of the main steps of the article loading method according to an embodiment of the present invention; Figure 2 This is a flowchart of the intelligent loading method for items according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating a multi-stage grouping method according to an embodiment of the present invention; Figure 4 A flowchart of a 3D packing optimizer according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the main modules of an article loading device according to an embodiment of the present invention; Figure 6 This is an exemplary system architecture diagram in which embodiments of the present invention can be applied; Figure 7 This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0019] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0020] It should be noted that the technical solutions disclosed in this invention, regarding the collection, updating, analysis, processing, use, transmission, and storage of user personal information, all comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.

[0021] In maritime shipping scenarios, it's common for shippers' cargo to be insufficient to fill an entire container. In such cases, to ensure normal transport and minimize space waste, it's necessary to consolidate cargo with other shippers' goods to fill the container. Currently, this consolidation operation primarily relies on manual screening, loading cargo from the same port of loading, destination, and time window into the same container for easier unloading. However, consolidation requires consideration of numerous factors, including cargo compatibility, container weight limits, and cargo placement. Manual consolidation struggles to account for all these factors, leading to poor consolidation results. Furthermore, manual consolidation is time-consuming and inefficient. To address these shortcomings, this invention proposes a cargo loading method that significantly reduces the time required to generate loading plans in consolidation scenarios and improves the accuracy of the generated plans.

[0022] Figure 1 This is a schematic diagram illustrating the main steps of a method for loading articles according to an embodiment of the present invention. Figure 1 As shown, the method for loading items mainly includes steps S101 to S103.

[0023] Step S101: Construct a tree data structure for the order data to be processed based on the three-dimensional index of the order origin, destination and loading time window. The order data to be processed is stored in the leaf nodes of the tree data structure, and the root node stores the order data to be processed in the current batch.

[0024] First, the system retrieves the pending order data provided by the user and extracts the order origin, destination, and loading time window for each order. The origin refers to the port where the order needs to be loaded, the destination is the port where the order will be unloaded, and the loading time window is the time range within which loading can take place at the origin. Then, a tree structure is constructed based on the extracted parameters to create a three-dimensional index, facilitating subsequent queries of the pending order data. Specifically, based on the root node, a tree structure is built using the origin as the first-dimensional index, the destination as the second-dimensional index, and the loading time window as the third-dimensional index, and the data for each pending order is stored in the leaf nodes.

[0025] In one embodiment, the root node of the current tree structure stores data for all pending orders in the current batch. When new order data is received, it can be determined whether to add the new order data to the current tree structure based on actual needs. If necessary, the new order data is stored in the corresponding leaf node based on the order's origin, destination, and loading time window. This leaf node can be an existing node or a newly created node. The determination based on actual needs can include factors such as vessel capacity and shipping time range.

[0026] In one embodiment, the acquired order data to be processed needs to be preprocessed before a tree-structured data structure is constructed. Preprocessing includes cleaning, standardization, and enhancement. Data cleaning: Identifying and correcting errors, missing values, and outliers in the order data, such as standardizing port names and unifying time formats. Data standardization: Converting order data from different sources and in different formats into a unified standard format. Specifically, this includes: standardizing item dimensions (length × width × height in meters); standardizing weight (in kilograms); standardizing port codes (UN / LOCODE international port codes); and standardizing dangerous goods information (UN international dangerous goods classification codes). Data enhancement: Calculating and adding enhanced features based on the original order data, including volume-to-weight ratio (the ratio of item volume to weight, used to determine item density); urgency (an urgency index calculated based on delivery time); customer priority (determined based on historical order volume and value); and loading complexity (a loading difficulty index assessed based on item characteristics). The preprocessed orders enable the rapid construction of a tree-structured data structure.

[0027] Step S102: Query the pending order data in the leaf nodes according to the target parameters corresponding to the pending order data, and generate order groups according to the query results corresponding to each target parameter.

[0028] For the constructed tree-structured data, target parameters can be obtained based on the order data to be processed, used for order grouping. Specifically, the target parameters can be parameters within the range of the order data to be processed, such as: origin point = port a, destination point = port b, and loading time window including 2025-10-04. Based on these parameters, the three-dimensional index in the tree-structured data is queried to determine the final matching leaf node, and the order data to be processed in the leaf node is retrieved. Different target parameters result in different order groups; for example, result a belongs to group 1 if target parameter a is used, and result b belongs to group 2 if target parameter b is used, etc. This target parameter enables coarse grouping of the order data to be processed based on the spatiotemporal dimension, obtaining various order groups for further fine-grained grouping operations. The R-tree spatial structure indexing algorithm enables fast queries with O(log n) time complexity. Taking the target parameter mentioned above as an example, firstly, it is determined whether there is an intersection with the target parameter based on the three-dimensional parameters of each child node stored in the root node. Child nodes that do not intersect are excluded. Then, it is verified whether the pending order data stored in the leaf node corresponding to the child node completely matches the target parameter. If they match, it is determined that the pending order data stored in the leaf node belongs to the order group corresponding to the target parameter.

[0029] In one embodiment, the target parameter is generated based on the parameter range of all orders in the order data to be processed, including both time and spatial ranges. The purpose of the target parameter is to group the data in each leaf node, rather than a simple query operation. Therefore, the target parameter needs to be generated within the parameter range of the orders in the order data to be processed, and the target parameter range includes the parameter range of the orders to ensure that all orders can be assigned to the corresponding order group.

[0030] In one embodiment, to avoid overly granular order grouping with a very small number of orders in each group, fuzzy time window matching is supported during queries, allowing for a loading time deviation of ±2 days. However, when the number of orders is large, the fuzzy time window matching rule can be disabled, depending on the actual situation. Specifically, for a target parameter starting point = port a, target point = port b, and a loading time window including 2025-10-04, the loading time window will be expanded to 2025-10-02 to 2025-10-06 during queries.

[0031] Step S103: Generate a loading scheme for orders to be processed based on loading constraints and order grouping.

[0032] After grouping the order data to be processed, a loading plan for each order group can be generated based on the order data and loading constraints within each group. These loading plans are then combined to generate the final, complete loading plan for all orders to be processed. Generating the loading plan for each order group requires solving for various pre-set loading constraints, and the optimal solution is used to generate the loading plan. These constraints include item compatibility constraints, item rotation and stacking constraints, and heavy-load bottom-level constraints.

[0033] The item loading method provided by the embodiments of the present invention can construct a tree data structure based on a three-dimensional index determined from the acquired order data to be processed, and achieve fast index query through the tree data structure to obtain order groups. This reduces the time complexity of the query operation from the traditional O(n²) to O(nlogn), significantly improving query speed and obtaining accurate query results. The order groups generated based on the query results, combined with preset loading constraints, help generate accurate loading schemes for the orders to be processed, greatly improving the efficiency of loading scheme generation.

[0034] In one embodiment, before constructing a tree-shaped data structure for the order data to be processed based on a three-dimensional index of the order's origin, destination, and loading time window, the method further includes: generating enhanced feature data of items in the order based on the original data of the order to be processed; and generating the order data to be processed based on the enhanced feature data and the original data of the order to be processed.

[0035] Before constructing the tree-structured data, to obtain more accurate grouping results and generate more accurate loading plans during subsequent order grouping, enhanced feature data can be further generated based on the original data of the orders to be processed. This enhanced feature data serves as parameters for grouping and participates in the calculation of the subsequent loading plan generation process. Specifically, enhanced feature data may include volume-to-weight ratio, i.e., the ratio of item volume to weight, which can be used to determine item density. When generating loading plans, items with higher density are prioritized and placed at the bottom. In addition, enhanced feature data may also include order priority, customer priority, etc., used to determine the order group to which the order data to be processed belongs. The generated enhanced feature data is merged with the original data of the orders to be processed to obtain the order data to be processed.

[0036] The item loading method provided by the embodiments of the present invention can further generate enhanced feature data based on the original data of the order to be processed, thereby enhancing the order data. The enhanced data helps to provide more accurate condition constraints when solving the loading scheme in the subsequent process, so as to improve the accuracy of the subsequent loading scheme generation.

[0037] In one embodiment, generating a loading scheme for orders to be processed based on loading constraints and order groups includes: for each order group, performing secondary grouping of the orders in the order group based on the attribute data of the items in the order group and the item compatibility rules to obtain order subgroups in each order group; and generating a loading scheme for orders to be processed based on the order groups, order subgroups and loading constraints.

[0038] When generating loading plans, the obtained order groups need to be further refined to obtain order subgroups. Specifically, the attribute data of items in each order group is obtained, including item type, hazard level, environmental requirements, and other compatibility attributes. This attribute data is mainly used to refine the grouping of items in the order group to determine information such as item combination and placement. Based on the obtained attribute data, a compatibility matrix can be constructed to obtain a knowledge graph of the orders to be processed, so that the attribute data can be quickly queried when generating loading plans, helping to generate the final loading plan. Among them, the hazard level includes the nine major categories of UN hazardy goods and their subcategories; environmental requirements include temperature, humidity, ventilation, etc.; item type includes odor, contamination, fragility, etc. Item compatibility rules are used to indicate which items are compatible and can be placed together, and which items are incompatible and cannot be placed together. According to the item compatibility rules, the items in the order group are further divided to obtain each order subgroup. For example, item compatibility rules include allowing items with similar environmental requirements to be placed together and correspondingly belonging to the same order subgroup. Other compatibility rules might include prohibiting contaminated items from being placed with other food or clothing. By combining order groups, order subgroups, and loading constraints, the problem is solved to obtain the loading scheme for each order subgroup, thus determining the loading scheme for the entire order group. Finally, the loading schemes of all order groups are merged to obtain the loading scheme for the order to be processed.

[0039] In one embodiment, instead of performing secondary grouping of order groups, the final loading plan can be generated directly based on the order groups and loading constraints. Compared to the secondary grouping approach, generating the loading plan directly from the order groups takes less time but has lower accuracy, while generating the loading plan after secondary grouping takes longer but yields more accurate results.

[0040] The item loading method provided by the embodiments of the present invention can perform secondary grouping on the basis of coarse grouping based on spatiotemporal dimensions, generating more accurate grouping results. Specifically, the secondary grouping employs grouping rules based on item attributes and item compatibility, enabling more precise grouping of items based on their attributes and improving the accuracy of the loading scheme generation.

[0041] In one embodiment, generating a loading scheme for an order to be processed based on order groups, order subgroups, and loading constraints includes: performing a multi-objective genetic solution based on cargo rotation and stacking constraints based on order subgroups and loading constraints to obtain loading schemes for each order group; and determining the loading scheme for the order to be processed based on the loading schemes for each order group.

[0042] When generating loading plans, a genetic algorithm is used for optimization. This involves solving a multi-objective genetic algorithm based on cargo rotation and stacking constraints to obtain the optimal solution. The optimal solution is then used to generate loading plans for each order group. Specifically, a two-layer encoding method is used: an item ID (Identity document) sequence and a rotation state. The rotation state describes the placement orientation of the item during loading. For any item, its original dimensions are "length (L) × width (W) × height (H)," but it can rotate around the X, Y, and Z axes within the container (while satisfying constraints such as "this face upwards"). The rotation state records the correspondence between the rotated length, width, and height and the original dimensions. For example, original dimensions: L=2m, W=1m, H=0.5m (rectangular cargo, no rotation constraints). Rotatable states include: State 0: No rotation, dimensions remain L×W×H = 2×1×0.5; State 1: Rotate 90° around the Z-axis, length and width interchanged, dimensions W×L×H = 1×2×0.5; State 2: Rotate 90° around the X-axis, width and height interchanged, dimensions L×H×W = 2×0.5×1; State 3: Rotate 90° around the Y-axis, length and height interchanged, dimensions H×W×L = 0.5×1×2. Fitness function: f = α×space utilization + β×center of gravity balance - γ×constraint violation. A genetic algorithm is used to solve the problem based on the encoding method and fitness function: 1) Initialize the population: Randomly generate 100 packing schemes (each scheme corresponds to one "item ID + rotation state" sequence). 2) Calculate fitness: For each scheme, calculate space utilization, center of gravity balance, and constraint violation, and substitute them into the function to obtain the f value (the higher the f value, the better the scheme). 3) Selection Operation: Retain the top 30% of schemes with the highest f-values ​​(elite retention strategy), and eliminate the remaining 70%. 4) Crossover / Mutation: Perform "crossover" (swapping part of the item sequences between two schemes) and "mutation" (modifying the rotation state of a single scheme) on the retained schemes to generate new schemes, supplementing the pool to 100. 5) Iteration Termination: Repeat steps 2-4 until "the highest f-value of schemes does not improve for 5 consecutive generations" or "the number of iterations reaches 100 generations." At this point, output the scheme with the highest f-value as the optimal packing scheme. Use this optimal scheme as the loading scheme for the order subgroup, and merge them to generate the final loading scheme for the orders to be processed.

[0043] In one embodiment, before determining the loading scheme of the order to be processed based on the loading scheme of each order group, the method further includes: performing collision detection on the loading scheme of each order group based on the size data of the items in each order subgroup; and adjusting the loading scheme of each order group based on the collision detection results.

[0044] Before determining the optimal packing scheme as the loading scheme for the order subgroups, collision detection is required, and the loading scheme needs to be adjusted based on the detection results. To improve detection efficiency, an octree data structure can be used to divide the space and construct a bounding box hierarchy of items, enabling fast collision detection between items within the divided space. The bounding box hierarchy can be an AABB (Axis-Aligned Bounding Box) hierarchy. For items identified as colliding, the loading scheme needs further adjustment until no colliding items remain. Several adjustment strategies are employed when adjusting the loading scheme, as shown below.

[0045] Strategy 1: Rotation Adjustment (Prioritize this option, lowest cost). Logic: For colliding items (e.g., item A), switch to an "untried rotation state" (rotation constraints must be met), recalculate the dimensions, and check if the collision still occurs. Example: Item A (original dimensions 2×1×0.5, rotation state 1) collides with item B (dimensions 1.5×1×0.5) at the bottom layer. Switch A's rotation state to 2 (dimensions 2×0.5×1). At this point, A's width shrinks from 1m to 0.5m, and it no longer overlaps with B. Adjustment complete.

[0046] Strategy 2: Position Adjustment (executed when rotation is ineffective). Logic: Within the same layer, move the colliding item (e.g., item A) to an unoccupied space within the current layer that meets the constraints (e.g., the free area on the right), maintaining its rotation state. Constraints: After moving, the following conditions must be met: "heavy objects at the bottom, center of gravity balanced" (e.g., if A is a heavy object, it cannot be moved to an upper layer), and it must not collide with other items. Example: If item A still collides with B after rotation, check if there is a 1.2×1×0.5 free space on the right side of the current layer. Move A to this area, recalculate the center of gravity, and ensure that the layer's center of gravity offset does not exceed the limit.

[0047] Strategy 3: Layer Adjustment (executed when position adjustment is ineffective). Logic: Move the colliding item (e.g., item A) to an "empty space on the upper or lower layer," and re-plan the layering scheme (e.g., increase the height of the current layer, or reduce the height of the upper layer). Constraints: Ensure the target layer's load-bearing capacity (e.g., if item A weighs 200kg, the upper layer's maximum load-bearing capacity is 300kg), and do not affect items on other layers. Example: Item A has no suitable location on the bottom layer, but there is 2×1×0.5m of empty space on the upper layer with sufficient load-bearing capacity. Move item A to the upper layer, and simultaneously adjust the layer heights of the bottom and upper layers (lower layer height decreases from 0.5m to 0.4m, upper layer height increases from 0.5m to 0.6m), ensuring no significant decrease in overall space utilization.

[0048] Strategy 4: Group Adjustment (executed when the first three strategies fail). Logic: If a colliding item cannot be adjusted within the current group (e.g., insufficient remaining space in the group), the item is marked as "to be regrouped" and fed back to the intelligent matching engine for rematching "other groups with suitable space". Example: Item A is ineffective in all adjustment strategies within Group 1. The engine re-adds A to the candidate order pool and matches it to Group 2 (same port of loading and destination, remaining space 2.5 × 1 × 0.5), re-executing packing optimization.

[0049] In one embodiment, a multi-objective genetic solution based on cargo rotation and stacking constraints is performed based on order subgroups and loading constraints to obtain loading schemes for each order group. This includes: layering the loading space to obtain multi-layer loading spaces; layering the items in the order subgroups based on the multi-layer loading spaces, the weight data and size data of the items in the order subgroups to obtain the items loaded in each layer of loading space; and for each layer of loading space, performing a multi-objective genetic solution based on cargo rotation and stacking constraints based on the items loaded in each layer of loading space and the loading constraints to obtain loading schemes for each order group.

[0050] Before generating the solutions for each order group, the space within the container can be layered during problem-solving to obtain multi-layered loading spaces. Loading strategies are then set for each layer to load items within the order subgroups based on their weight and size data, as well as the specific loading strategies for each layer. These strategies include: horizontal layering: dividing the container space horizontally into multiple layers and loading layer by layer starting from the bottom; heavy-load bottom principle: prioritizing heavier goods at the bottom to ensure loading stability; maximizing area utilization: using a 2D (planar) packing algorithm to maximize the area utilization of each layer; and minimizing inter-layer gaps: dynamically adjusting layer height to reduce inter-layer gaps and improve space utilization. Based on these strategies, the items to be loaded in each layer are generated. Multi-objective genetic problem-solving is then performed using loading constraints, and the optimal solution corresponds to the loading scheme for the order group.

[0051] In one embodiment, the three core algorithms for generating loading schemes—layered loading strategy, genetic algorithm optimization, and real-time collision detection—are set in the same bin packing optimizer. By inputting order groups, order subgroups, and corresponding order data to be processed into the bin packing optimizer, the loading scheme of the order groups output by the bin packing optimizer can be obtained.

[0052] The item loading method provided by the embodiments of the present invention can perform precise binning optimization on the secondary grouping results, i.e., the generation of loading schemes. Accurate generation of loading schemes for items in different groups is achieved through three core algorithms: hierarchical strategy, genetic algorithm, and collision detection. Furthermore, the progressive design of different algorithms allows for the step-by-step solution of complex problems, reducing computational difficulty and improving computational efficiency. The step-by-step solution method also enhances the scalability and maintainability of the algorithm.

[0053] In one embodiment, generating order groups based on query results corresponding to each target parameter includes: determining at least one target parameter corresponding to each pending order based on query results corresponding to each target parameter; in response to a pending order having multiple target parameters, determining a first parameter corresponding to any pending order based on the timeliness information and corresponding user information of any pending order; in response to a pending order having only one target parameter, determining the target parameter corresponding to any pending order as the first parameter; and generating order groups based on the first parameter corresponding to each pending order.

[0054] In some cases, when querying pending orders based on target parameters, the same order may match multiple target parameters. In such cases, it is necessary to select one target parameter as the matching target parameter for that order. Specifically, this selection can be based on the order's urgency information and user information, where urgency information can be the order's urgency and user information can be the user's level. The order's score is obtained by weighted summing of these two factors, and the first target parameter is determined based on this score. Orders are then grouped according to this first parameter. Different target parameters correspond to different order grouping urgency levels; for example, they might correspond to order groups for 10.24 and 10.25 respectively. When an order has a higher score, its priority is higher, and it needs to be assigned to the faster-urgent order group, i.e., the 10.24 order group.

[0055] In one embodiment, when calculating the order score, in addition to failure information and user information, profit margin and loading complexity parameters may also be included. The profit margin is the profit margin that a loading order can obtain based on pricing rules. Therefore, the formula for calculating the order score S is S = ×Urgency+ ×Customer Level+ ×Profit Margin+ × Loading complexity. , , and The pre-set weight values ​​can be dynamically adjusted based on business feedback and matching results.

[0056] In one embodiment, the loading scheme for pending orders includes multiple schemes, and the method further includes: displaying multiple pending order loading schemes in a three-dimensional display manner, so that the user can select a target pending order loading scheme based on the multiple pending order loading schemes.

[0057] In one embodiment, the loading scheme obtained through the solution process can be multiple schemes that satisfy the constraints. In this case, instead of selecting the optimal solution as the final scheme, multiple relatively good schemes that satisfy the constraints are selected. When selecting the optimal solution, the objective can be minimizing space occupation, achieving optimal space utilization, or a combination of multiple objectives. When generating multiple schemes, for example, an optimal range for space utilization can be set, and multiple loading schemes within that range can be obtained. These multiple loading schemes can be displayed to the user in a highly intuitive 3D way, allowing the user to choose the appropriate loading scheme based on the visual effect. The main display functions include: (1) 3D (three-dimensional) container loading effect display: intuitively display the placement and order of goods in the container; (2) multi-view observation: support multiple view operations such as rotation, zoom, and cross-section viewing; (3) constraint visualization: use different colors to mark the dangerous goods area, center of gravity position, rotation restrictions and other constraint information; (4) loading sequence animation: simulate the animation effect of the actual loading process to guide on-site operation; (5) scheme comparison: support the side-by-side comparison of multiple container loading schemes to assist in decision-making on the optimal scheme.

[0058] The item loading method provided by this invention can accurately group order data across different dimensions during the order processing stage by combining coarse and fine grouping, resulting in accurate grouping results and improving the accuracy of subsequent loading schemes. Specifically, a three-dimensional index structure is introduced in the coarse grouping stage to enable fast querying, significantly reducing time complexity and increasing query speed. In the fine grouping stage, a constructed compatibility knowledge matrix further improves grouping efficiency, and dynamic grouping scoring balances the impact of various business factors, further optimizing the grouping results. For the obtained grouping results, a hierarchical strategy, genetic algorithm, and collision detection are used to progressively solve complex problems, reducing the difficulty of the solution and improving computational efficiency. Finally, an accurate loading scheme can be obtained based on the solution results.

[0059] like Figure 2 The diagram shows the main flowchart of the intelligent loading method for items according to an embodiment of the present invention. The method mainly involves four structures: an order preprocessing unit, an intelligent matching engine, a 3D packing optimizer, and a constraint checker. The order preprocessing unit performs data cleaning, standardization, and data augmentation operations on the received order data; the intelligent matching engine performs spatiotemporal matching of orders, compatibility checks, and calculates order weight scores; the 3D packing optimizer performs layered loading, solves the problem using a genetic algorithm, and performs collision detection; and the constraint checker performs hazardous materials inspection, restricts item rotation, and checks item weight distribution.

[0060] like Figure 3 The diagram illustrates a multi-stage grouping method according to an embodiment of the present invention. The acquired order data to be processed is first coarsely segmented in the first stage, then further subdivided in the second stage, and finally the refined groupings are output to the subsequent 3D packing optimizer. The first stage is a spatiotemporal matching stage, including spatiotemporal matching based on a constructed tree-like data structure and supporting fuzzy matching strategies to obtain coarsely segmented results. The second stage is a precise matching stage based on attribute dimensions, including constructing an item compatibility knowledge matrix, dynamically calculating order scores, and performing attribute dimension matching to obtain subdivided results.

[0061] like Figure 4 The diagram shows a flowchart of a 3D packing optimizer according to an embodiment of the present invention. Order data from the segmentation results is preprocessed and then calculated using the core optimization algorithm layer. Preprocessing includes item size standardization and verification, and weight sorting and classification. The core optimization algorithm layer includes a hierarchical loading strategy algorithm, a multi-objective genetic algorithm, and a real-time collision detection engine. The calculation results obtained from the core optimization algorithm are input to the constraint processing layer for constraint adjustment and verification. The output of the constraint processing layer can be further analyzed through the result output layer, including scheme evaluation, iterative optimization control, optimal solution selection, scheme performance analysis, and scheme quality verification, ultimately yielding the loading scheme for the order to be processed. The obtained loading scheme can be visualized, and a performance report of the loading scheme is provided. The performance report may include data such as space utilization rate and constraint satisfaction rate.

[0062] Figure 5 This is a schematic diagram of the main modules of an item loading device according to an embodiment of the present invention. As shown in the figure, the item loading device 500 mainly includes a construction module 501, a query module 502, and a generation module 503.

[0063] Module 501 is used to construct a tree data structure of the order data to be processed based on the three-dimensional index of the order starting point, target point and loading time window. The order data to be processed is stored in the leaf nodes of the tree data structure, and the root node stores the order data to be processed in the current batch. The query module 502 is used to query the pending order data in the leaf nodes according to the target parameters corresponding to the pending order data, and generate order groups according to the query results corresponding to each target parameter; The generation module 503 is used to generate a loading scheme for orders to be processed based on loading constraints and order grouping.

[0064] The item loading apparatus provided in this embodiment of the invention can construct a tree data structure based on a three-dimensional index determined from the acquired order data to be processed, and achieve fast index query through the tree data structure to obtain order groups. This reduces the time complexity of the query operation from the traditional O(n²) to O(nlogn), significantly improving query speed and obtaining accurate query results. The order groups generated based on the query results, combined with preset loading constraints, help generate accurate loading schemes for the orders to be processed, greatly improving the efficiency of loading scheme generation.

[0065] In one embodiment, the generation module 503 is further configured to: generate enhanced feature data of items in the order to be processed based on the original data of the order to be processed; and generate order data to be processed based on the enhanced feature data and the original data of the order to be processed.

[0066] In one embodiment, the generation module 503 is further configured to: for each order group, perform secondary grouping of the orders in the order group based on the attribute data of the items in the order group and the item compatibility rules to obtain order subgroups in each order group; and generate a loading scheme for the orders to be processed based on the order group, the order subgroups and the loading constraints.

[0067] In one embodiment, the generation module 503 is further configured to: perform a multi-objective genetic solution based on cargo rotation and stacking constraints based on order subgroups and loading constraints to obtain loading schemes for each order group; and determine loading schemes for orders to be processed based on loading schemes for each order group.

[0068] In one embodiment, the item loading device 500 further includes: a detection module 504 (not shown in the figure), used to perform collision detection on the loading scheme of each order group based on the size data of the items in each order subgroup; and an adjustment module 505 (not shown in the figure), used to adjust the loading scheme of each order group based on the collision detection results.

[0069] In one embodiment, the generation module 503 is further configured to: perform layered processing on the loading space to obtain multi-layered loading spaces; perform layered loading on the items in the order subgroups based on the weight data and size data of the items in the multi-layered loading spaces and the order subgroups to obtain the items loaded in each layer of loading space; and for each layer of loading space, perform multi-objective genetic solution based on cargo rotation and stacking constraints based on the items loaded in each layer of loading space and the loading constraints to obtain the loading scheme for each order group.

[0070] In one embodiment, the generation module 503 is further configured to: determine at least one target parameter corresponding to each pending order based on the query results corresponding to each target parameter; in response to the number of target parameters corresponding to any pending order being multiple, determine a first parameter corresponding to any pending order based on the timeliness information of any pending order and the corresponding user information; in response to the number of target parameters corresponding to any pending order being one, determine the target parameter corresponding to any pending order as the first parameter; and generate an order group based on the first parameter corresponding to each pending order.

[0071] In one embodiment, the loading scheme for the pending orders includes multiple schemes, and the item loading device 500 further includes: a display module 506 (not shown in the figure), used to display multiple pending order loading schemes in a three-dimensional display manner, so that the user can select the target pending order loading scheme based on the multiple pending order loading schemes.

[0072] The item loading apparatus provided in this embodiment of the invention can accurately group order data across different dimensions during the order processing stage by combining coarse grouping and fine grouping, resulting in accurate grouping results and improving the accuracy of subsequent loading schemes. Specifically, a three-dimensional index structure is introduced in the coarse grouping stage to enable fast querying, significantly reducing time complexity and increasing query speed. In the fine grouping stage, a constructed compatibility knowledge matrix further improves grouping efficiency, while dynamic grouping scoring balances the impact of various business factors, further optimizing the grouping results. For the obtained grouping results, a hierarchical strategy, genetic algorithm, and collision detection are used to progressively solve complex problems, reducing the difficulty of the solution and improving computational efficiency. Finally, an accurate loading scheme can be obtained based on the solution results.

[0073] Figure 6 An exemplary system architecture 600 is shown, in which the method or apparatus for loading items according to embodiments of the present invention can be applied.

[0074] like Figure 6As shown, system architecture 600 may include terminal devices 601, 602, and 603, a network 604, and a server 605. Network 604 serves as the medium for providing communication links between terminal devices 601, 602, and 603 and server 605. Network 604 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0075] Users can use terminal devices 601, 602, and 603 to interact with server 605 via network 604 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 601, 602, and 603, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0076] Terminal devices 601, 602, and 603 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0077] Server 605 can be a server that provides various services, such as a backend management server (for example only) that supports shopping websites browsed by users using terminal devices 601, 602, and 603. The backend management server can analyze and process data such as received loading scheme generation requests, and feed back the processing results (e.g., loading schemes - for example only) to the terminal devices.

[0078] It should be noted that the item loading method provided in this embodiment of the invention is generally executed by server 605, and correspondingly, the item loading device is generally located in server 605.

[0079] It should be understood that Figure 6 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0080] The following is for reference. Figure 7 It shows a schematic diagram of the structure of a computer system 700 suitable for implementing terminal devices or servers of the present invention. Figure 7 The terminal device or server shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0081] like Figure 7As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 702 or programs loaded from storage section 708 into random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the system 700. The CPU 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0082] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.

[0083] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by central processing unit (CPU) 701, it performs the functions defined above in the system of this invention.

[0084] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0085] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0086] The units or modules described in the embodiments of the present invention can be implemented in software or hardware. The described units or modules can also be housed in a processor; for example, a processor can be described as including a construction module, a query module, and a generation module. The names of these units or modules do not necessarily limit the specific unit or module itself; for example, a sending unit can also be described as "a unit that sends an image retrieval request to a connected server."

[0087] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to include: A tree-like data structure for the order data to be processed is constructed based on a three-dimensional index of the order origin, destination, and loading time window. The order data to be processed is stored in the leaf nodes of the tree data structure, and the root node stores the order data to be processed in the current batch. Query the pending order data in the leaf nodes according to the target parameters corresponding to the pending order data, and generate order groups according to the query results corresponding to each target parameter; A loading scheme is generated based on loading constraints and order grouping to produce orders to be processed.

[0088] According to the technical solution of this invention, a tree-like data structure can be constructed based on the three-dimensional index determined by the acquired order data to be processed. This tree-like data structure enables fast index querying to obtain order groups, reducing the time complexity of the query operation from the traditional O(n²) to O(nlogn), significantly improving query speed and obtaining accurate query results. The order groups generated based on the query results, combined with preset loading constraints, help generate accurate loading schemes for the orders to be processed, greatly improving the efficiency of loading scheme generation.

[0089] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for loading articles, characterized in that, include: A tree-like data structure for the order data to be processed is constructed based on a three-dimensional index of the order origin, destination, and loading time window. The order data to be processed is stored in the leaf nodes of the tree-like data structure, and the root node stores the order data to be processed for the current batch. Based on the target parameters corresponding to the order data to be processed, query the order data to be processed in the leaf node, and generate order groups based on the query results corresponding to each target parameter; A loading scheme for orders to be processed is generated based on loading constraints and the order grouping.

2. The method according to claim 1, characterized in that, Before constructing the tree data structure of the order data to be processed based on the three-dimensional index of the order origin, destination, and loading time window, the method further includes: Enhanced feature data of the items in the pending orders are generated based on the raw data of the pending orders; The pending order data is generated based on the enhanced feature data and the original data of the pending orders.

3. The method according to claim 1 or 2, characterized in that, The loading scheme for generating pending orders based on loading constraints and order grouping includes: For each order group, the orders in the order group are further grouped based on the attribute data of the items in the order group and the item compatibility rules to obtain the order subgroups in each order group; A loading scheme for the orders to be processed is generated based on the order grouping, the order subgrouping, and the loading constraints.

4. The method according to claim 3, characterized in that, The loading scheme for generating orders to be processed based on the order grouping, the order subgrouping, and the loading constraints includes: Based on the order subgroups and the loading constraints, a multi-objective genetic solution based on cargo rotation and stacking constraints is performed to obtain the loading scheme for each order group; The loading scheme for the orders to be processed is determined based on the loading scheme of each order group.

5. The method according to claim 4, characterized in that, Before determining the loading scheme for the order to be processed based on the loading scheme of each order group, the method further includes: Collision detection is performed on the loading scheme of each order group based on the size data of the items in each order subgroup; The loading scheme for each order group is adjusted based on the collision detection results.

6. The method according to claim 4 or 5, characterized in that, The multi-objective genetic solution based on cargo rotation and stacking constraints, performed on the order subgroups and loading constraints, yields loading schemes for each order group, including: The loading space is divided into layers to obtain multi-layer loading space; Based on the weight and size data of the items in the multi-layer loading space and the order subgroup, the items in the order subgroup are loaded in layers to obtain the items loaded in each layer of loading space; For each loading space, a multi-objective genetic solution based on cargo rotation and stacking constraints is performed based on the items loaded in each loading space and the loading constraints to obtain the loading scheme for each order group.

7. The method according to claim 1 or 2, characterized in that, The step of generating order groups based on the query results corresponding to each target parameter includes: Based on the query results corresponding to each target parameter, determine at least one target parameter for each order to be processed; In response to the fact that there are multiple target parameters corresponding to any pending order, the first parameter corresponding to any pending order is determined based on the timeliness information of the pending order and the corresponding user information; In response to the fact that the number of target parameters corresponding to any pending order is one, the target parameter corresponding to any pending order is determined as the first parameter; Order groups are generated based on the first parameter corresponding to each pending order.

8. The method according to claim 1 or 2, characterized in that, The loading scheme for the orders to be processed includes multiple schemes, and the method further includes: The multiple order loading schemes are displayed in a three-dimensional way, allowing users to select the target order loading scheme based on these schemes.

9. A device for loading articles, characterized in that, include: The construction module is used to construct a tree data structure of the order data to be processed based on the three-dimensional index of the order starting point, target point and loading time window. The order data to be processed is stored in the leaf nodes of the tree data structure, and the root node stores the order data to be processed in the current batch. The query module is used to query the pending order data in the leaf node according to the target order parameters corresponding to the pending order data, and generate order groups according to the query results corresponding to each target parameter; The generation module is used to generate a loading scheme for orders to be processed based on loading constraints and the order grouping.

10. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-8.

11. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

Images