Visual modeling method and system for warehouse products

Abstract

The invention relates to the technical field of warehouse management, in particular to a visual modeling method and system for warehouse products, and the method is characterized in that the method comprises the following steps: S1, obtaining order data of products; s2, preprocessing the order data to generate net data; s3, constructing a grouping processing module, and grouping the products by the grouping processing module based on the net data to generate a product packet; s4, constructing a stacking module, and performing stacking calculation on the products in the product package by the stacking module based on the net data and the product package to generate a stacking scheme of the product package; and S5, a three-dimensional modeling module is constructed, the three-dimensional modeling module carries out three-dimensional modeling on the product package based on the stacking scheme, and a visual three-dimensional model is generated. The method has the advantages of being independent of manpower, adaptive to diversified orders and capable of automatically generating the visual model.

Description

Technical Field

[0001] This invention relates to the field of warehouse management technology, and more specifically, to a visual modeling method and system for warehouse products. Background Technology

[0002] In warehouse management within the construction industry, product stacking is a critical operation. Due to the wide variety of products, significant differences in size and weight, and the need to meet multiple constraints such as space utilization, stacking stability, and transportation limitations, manually developing solutions often proves unsatisfactory, time-consuming, labor-intensive, and highly dependent on the experience level of the operators. This results in inconsistent solution quality, making it difficult to adapt to complex and ever-changing product order demands.

[0003] In terms of visualization, traditional methods typically only provide simple two-dimensional drawings or text descriptions.

[0004] Further approaches involve preprocessing order data using computer algorithms to optimize grouping rules and exploring the use of 3D modeling tools to display stacking results. However, these solutions still have limitations in practical applications: First, the data processing is typically tailored to specific types of product orders, lacking versatility and failing to flexibly address diverse order needs; second, existing 3D visualization tools are mostly static displays, making it difficult to achieve dynamic interactive functions, such as viewing stacking details layer by layer, selecting specific products, or displaying the stacking process by level; third, existing technologies lack a complete integrated chain from order input to solution generation to visualization, requiring operators to switch between multiple systems, impacting operational efficiency. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a visualization modeling method and system for warehouse products, which has the advantages of not relying on manual labor, adapting to diverse orders, and automatically generating visualization models.

[0006] The above-mentioned technical objective of the present invention is achieved through the following technical solution: a visual modeling method and system for warehouse products, comprising the following steps:

[0007] S1. Obtain product order data;

[0008] S2. Preprocess the order data to generate net data;

[0009] S3. Construct a grouping processing module, which groups the products based on the net data to generate product packages;

[0010] S4. Construct a stacking module. The stacking module performs stacking calculations on the products within the product package based on the net data and the product package, and generates a stacking scheme for the product package.

[0011] S5. Construct a 3D modeling module, which performs 3D modeling of the product package based on the stacking scheme to generate a visual 3D model.

[0012] In one embodiment, the preprocessing includes data verification and data cleaning.

[0013] In one embodiment, the operation steps of the group processing module are as follows:

[0014] S31. Obtain the size, weight, and shape of the product based on the net data;

[0015] S32. Based on the size, weight, and shape, the products are grouped using a grouping algorithm to generate product packages.

[0016] In one embodiment, the operation steps of the stacking module are as follows:

[0017] S41. Obtain the size, weight, and shape of the product based on the net data;

[0018] S42. Based on the size, weight and shape, a stacking optimization algorithm is used to perform stacking calculations on the products in the product package to generate the stacking scheme.

[0019] In one embodiment, the content displayed by the visualized 3D model includes:

[0020] The dimensions, weight, and total number of products of the visualized 3D model; the number and arrangement of products in each layer of the visualized 3D model; the process of stacking the products to form the visualized 3D model; the overall frame outline of the visualized 3D model in a semi-transparent state; and the different display effects of selected products in the visualized 3D model compared to unselected products.

[0021] A visualization modeling system for warehouse products, implementing the aforementioned visualization modeling method for warehouse products, includes:

[0022] At the warehouse end, based on the information of the items, information is entered into the items and item data is generated;

[0023] The PC terminal is used to receive the item data, model the item, and generate a visual model.

[0024] The warehouse terminal and the PC terminal communicate via a network.

[0025] In one embodiment, the PC is equipped with an AI model to provide algorithms and apply them to the modeling.

[0026] The above-mentioned visualization modeling method and system for warehouse products has the following beneficial effects:

[0027] Breaking away from reliance on human experience, it can automatically adapt to diverse order requirements, significantly improving the efficiency and quality of stacking solution generation. With the help of intuitive 3D visualization models and dynamic interactive functions, operators can quickly and accurately grasp the details of the solution and execute operations efficiently, which not only effectively reduces rework, but also promotes the overall improvement of warehouse operation efficiency in the construction industry. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating the visualization modeling method in this embodiment;

[0029] Figure 2 This is a schematic diagram of the steps of the visualization modeling method in this embodiment. Detailed Implementation

[0030] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0031] A visual modeling method for warehouse products, such as Figure 1-2 As shown, it includes the following steps:

[0032] S1. Obtain product order data;

[0033] S2. Preprocess the order data to generate net data;

[0034] Preprocessing includes data validation and data cleaning;

[0035] In this invention, the product order data is first acquired, and then preprocessed by data verification and data cleaning to generate net data. The purpose of this step is to acquire parameters in the product order, such as product size, weight, quantity, etc., to prepare for the subsequent determination of product grouping rules, picking and placing order, and stacking method within each group.

[0036] Specifically, the process begins by receiving product order data, which includes parameters such as product dimensions, weight, and quantity. The order data is first validated to check its completeness and consistency, ensuring, for example, that there are no missing or outlier values ​​for dimensions and weight. Subsequently, the data is cleaned to remove redundant or erroneous information, ensuring the accuracy of subsequent processing steps. This systematic data processing ensures the reliability of the input data, laying the foundation for generating an efficient stacking solution.

[0037] S3. Construct a grouping processing module. The grouping processing module groups products based on net data and generates product packages.

[0038] The specific steps for the group processing module to operate are as follows:

[0039] S31. Obtain the product's dimensions, weight, and shape based on net data;

[0040] S32. Based on size, weight, and shape, a grouping algorithm is used to group the products and generate product packages;

[0041] To identify and group a variety of products, this invention employs a generalized grouping algorithm to automatically group products (i.e., form multiple product packs). This algorithm can handle various data types and is not bound to fixed grouping logic, making logical judgments based on specific circumstances. The algorithm comprehensively considers the product's size, weight, shape, and space utilization and stability requirements during stacking, dynamically adjusting grouping rules to adapt to the diverse needs of different orders. The grouping process maximizes warehouse space utilization through mathematical optimization principles while ensuring that the internal structure of each product pack facilitates subsequent stacking.

[0042] S4. Construct a stacking module. Based on net data and product packages, the stacking module performs stacking calculations on the products within the product packages to generate a stacking scheme for the product packages.

[0043] The specific steps for operating the stacking module are as follows:

[0044] S41. Obtain the product's dimensions, weight, and shape based on net data;

[0045] S42. Based on size, weight and shape, use a stacking optimization algorithm to perform stacking calculations on the products in the product package and generate a stacking scheme;

[0046] In this invention, a stacking optimization algorithm is used to automatically generate a stacking scheme for each product package. This algorithm determines the placement order and spatial layout of the products within the package based on their size, weight, and grouping rules. The algorithm is designed to be versatile, dynamically adjusting stacking rules according to changes in product characteristics. For example, it prioritizes placing heavier items at the bottom, lighter items on top, or using specific shaped products in a nesting configuration to ensure stacking stability and space efficiency. The generated stacking scheme includes the length, width, height, total weight of each product package, and the number and arrangement of products in each layer.

[0047] In this invention, the algorithm in the module can be reliably preset in advance, or it can be provided by other modules (such as AI modules) (the modules of this invention obtain the algorithm through communication connections, etc.), and then applied to grouping and stacking.

[0048] S5. Construct a 3D modeling module. The 3D modeling module performs 3D modeling of the product package based on the stacking scheme and generates a visual 3D model.

[0049] The content displayed in the visualized 3D model includes:

[0050] The visualization includes the size, weight, and total number of products in the 3D model; the number and arrangement of products on each layer of the 3D model; the process of stacking products into the 3D model; the overall frame outline of the 3D model in a semi-transparent state; and the difference in display effect between selected and unselected products in the 3D model.

[0051] To better illustrate the practical application of this invention in scenarios involving single-product orders (i.e., a single product package), multi-product grouped orders (i.e., multiple product packages), and stacking optimization algorithms, this invention is described in detail through three embodiments.

[0052] Example 1

[0053] In the case of a single product order, the generated stacking plan is transferred to the 3D modeling module to construct a three-dimensional model of the product package. The model displays the overall structure of the product, including length, width, height, total weight, and product distribution. Operators can perform the following operations through the interactive interface: view overall information of the product package and annotate dimensions; click to view detailed data of the product package; search the model by specific size and single size; display the stacking process step by step; and display the internal structure in a semi-transparent mode to help operators understand the placement order of each layer of products.

[0054] Example 2

[0055] In complex order scenarios involving multiple product types and grouping requirements, the generated stacking scheme is transferred to the 3D modeling module to construct a 3D model of the product package. Taking the generated First Floor and Ground Floor of the product package as an example, it supports switching between them to view the overall size, weight, and product distribution of each package. It supports rotating the model or switching to side view, top view, and other perspectives to observe the spatial structure of the product package; it displays the stacking order in the First Floor of the package step by step, with a transparent mode highlighting the internal interlocking details; it highlights the plasterboard (3600mm*1200mm*13mm) in the First Floor and allows filtering its distribution position in the First Floor of the package.

[0056] Example 3

[0057] This embodiment focuses on a single-group stacking scenario for gypsum board orders in a construction industry warehouse, illustrating the system's technical implementation in product stacking sorting and 3D visualization model generation. This embodiment primarily describes the stacking algorithm rules for gypsum board products.

[0058] The first step is to input the order data, which includes the type, size, weight, and quantity of plasterboard products. For example, an order might include:

[0059] 20 pieces of gypsum board A (6000mm×1200mm×12mm, 20kg / piece);

[0060] 15 pieces of gypsum board B (3000mm×1200mm×12mm, 10kg / piece);

[0061] 15 pieces of gypsum board C (6000mm×1350mm×12mm, 22kg / piece);

[0062] 10 pieces of gypsum board D (3000mm×1350mm×12mm, 11kg / piece);

[0063] By validating order data, we ensure that size and weight parameters are complete and free of anomalies (such as exceeding limits), and remove redundant information (such as duplicate entries). This step provides an accurate data foundation for the stacking algorithm.

[0064] The second step is to generate a single product package for the single-group scenario, containing all 60 plasterboards. Grouping takes into account truck space (or other storage space) constraints (length ≤ 6000mm, width ≤ 1500mm, height ≤ 1000mm) and loading stability to ensure maximum space utilization.

[0065] The third step is to generate a stacking scheme. The stacking algorithm is designed with general rules for gypsum board products, optimizing size constraints, stability, and special sorting requirements, as detailed below:

[0066] Regarding stacking rules, size restrictions apply: product packages must not exceed 6000mm in length, 1500mm in width, and 1000mm in height to meet truck loading requirements.

[0067] Stability and space utilization: Adopting the pyramid principle, the width of the bottom plate is greater than or equal to that of the top plate, ensuring loading stability and storage accessibility, while optimizing space utilization.

[0068] Length splicing: If the number of similar boards is odd or splicing is required, prioritize splicing two boards with a length ≤ 3000mm to form a 6000mm length, creating a complete layer. If an odd number of boards exist, the system matches other boards with odd numbers, prioritizing boards with the closest height for splicing; if the heights are the same, further select the combination with the closest spliced ​​length to 6000mm. The spliced ​​board with a length of 6000mm is placed on top of the unspliced ​​boards.

[0069] Special sorting: 1350mm wide boards (C, D) should be clustered in the middle layer as much as possible, avoiding placement at the bottom; 1200mm wide boards (A, B) should be placed at the bottom layer first. If an order contains both 1200mm and 1350mm wide boards, the algorithm ensures that the 1350mm boards are clustered in the middle and the 1200mm boards are placed at the bottom. Meanwhile, non-joinable products are stacked according to a pyramid principle, ensuring that the narrower boards are placed at the top.

[0070] Based on the above orders, the algorithm generates the following stacking scheme:

[0071] First layer: 20 6000×1200mm panels (A), arranged in a single layer, each panel measuring 6000×1200×12mm, total weight 400kg. A 1200mm wide panel is placed at the bottom to provide a stable base, conforming to the pyramid principle. Second layer: 14 3000×1200mm panels (B), two panels joined together along the length to form a 6000mm section, one set (2 panels) per layer, measuring 6000×1200×12mm, total weight 140kg. One remaining B panel is reserved for later assembly. Third layer: 15 6000×1350mm panels (C), arranged in a single layer, measuring 6000×1350×12mm, weight 330kg. A 1350mm wide panel is placed in the middle layer to meet aggregation requirements. Fourth layer: 11 pieces of 3000×1350mm boards (D), two boards are spliced ​​together along the length to form a 6000mm section, one set (2 boards) per layer, dimensions 6000×1350×12mm, weight 110kg. The 1350mm boards are kept clustered in the middle, with one remaining D board reserved for later splicing. Fifth layer: 1 piece of 3000×1200mm board (B, odd number remaining), spliced ​​with 1 piece of 3000×1350mm board (D, odd number remaining), dimensions 6000×1350×12mm, weight 21kg. The spliced ​​layer is placed above the unspliced ​​layers, and the splicing principle follows the rule of prioritizing the smallest height difference and ensuring the spliced ​​length is closest to 6000mm.

[0072] In this invention, the generation and interaction of the three-dimensional visualization model is as follows: the generated stacking scheme is transmitted to the three-dimensional modeling module to construct a visualized three-dimensional model of each product package. The model can display the overall structural information of the product package, including length, width, height, total weight, and the distribution of products in each layer. To facilitate understanding and operation by operators, this invention provides dynamic interactive functions, including: (1) Overall information viewing: (i.e., displaying the complete three-dimensional model of the product package), including overall size and weight information and the total number of products. (2) Layer-by-layer viewing: supports viewing stacking details by layer, displaying the number and placement of products in each layer. (3) Step-by-step stacking display: through the interactive interface, the step-by-step stacking process is displayed. The model can be switched to transparent mode, retaining only the frame outline, which is convenient for observing the internal structure. (4) Stacked product query: supports selecting products of a specific size or a type of product for highlighting.

[0073] The aforementioned 3D visualization function is achieved through computer graphics processing technology, based on spatial coordinate mapping and dynamic rendering principles, ensuring the intuitiveness and interactivity of the model. This technical solution constructs a closed-loop processing chain from order input to solution display through integrated data processing, group optimization, stacking planning, and 3D visualization functions. Compared with existing technologies, this invention does not rely on human experience, can automatically adapt to diverse order requirements, and significantly improves the efficiency and quality of stacking solution generation. Simultaneously, through intuitive 3D visualization models and dynamic interactive functions, operators can quickly understand the details of the solution and implement operations, thereby effectively reducing rework and improving the overall efficiency of warehouse operations in the construction industry.

[0074] Comparative example:

[0075] Compared with existing technologies, this invention demonstrates significant advantages in warehouse operations within the construction industry through intelligent data processing, stacking optimization algorithms, and dynamic 3D visualization technology. The advantages and beneficial effects of this invention are described below, prioritizing them from highest to lowest, and combining quantitative data comparisons and core technical features. The data is based on a six-month actual test conducted in 2024 at a building materials warehouse (processing 12,000 orders annually, storing gypsum board, steel bars, timber, etc., with order sizes ranging from 100 to 2,000 pieces).

[0076] This invention, through automated algorithms and 3D visualization technology, comprehensively outperforms traditional manual methods. The following quantitative comparisons of space occupancy, picking efficiency, solution generation time, order processing capacity, rework rate, and cost demonstrate its performance advantages.

[0077] Regarding space occupancy, as shown in Table 1, taking an order of 1000 gypsum boards (standard size range 1200mm×2400mm×12.5mm—1350mm×6000mm×13mm, average weight approximately 25kg / board) as an example, this system, through optimized algorithms, achieved an average space occupancy rate of 93.8%, while the manual solution averaged 86.2%, an improvement of 7.6%. Testing covered 100 orders (including standard and non-standard sizes). The system solution achieved a higher space occupancy rate than the manual solution in 95% of the orders, and consistently maintained above 90% in complex orders (including irregularly shaped timber and steel bars), while the manual solution fluctuated to below 80%.

[0078] Scheme type Average space occupancy rate (%) Increase (%) manual solution 86.2 - System Solution 93.8 7.6

[0079] Table 1

[0080] Regarding picking efficiency, as shown in Table 2, the algorithm significantly reduces the picking error rate through precise stacking order planning and 3D visualization guidance. In the test of 100 orders (50 to 100 pieces per order, including plasterboard and accessories of different sizes and weights), the manual picking method had an average picking error rate of 14.2% (approximately 14.2 adjustments needed per 100 picking attempts), mainly due to incorrect order or improper product selection. The system method reduced the error rate to 4.5% and achieved a success rate of 95.5%. For example, for one order (800 pieces of plasterboard and 200 accessories), the manual method resulted in 12 rework attempts due to incorrect order (each adjustment taking approximately 15 minutes), while the system method achieved zero rework through layer-by-layer guidance, saving 180 minutes of operation time.

[0081] Scheme type Average Pickup Error Rate (%) Success rate (%) manual solution 14.2 85.8 System Solution 4.5 95.5

[0082] Table 2

[0083] Regarding solution generation time, as shown in Table 3, the test used a conventional server (Intel Xeon E5-2630 v4 processor, 32GB memory, Ubuntu 22.04 system). For simple orders (10 products, 500 pieces, such as standard drywall), manual solution development by experienced workers took 8-10 minutes (average 9 minutes); for complex orders (50 products, 1500 pieces, including long strips and heat and sound insulation products), it took 15-25 minutes (average 20 minutes). This system generated the optimal solution and 3D model within 60 seconds for both types of orders, improving efficiency by approximately 92%. For example, for an order (1200 pieces, including 600 drywall panels, 400 long iron strips, and 200 accessories), manual solution development took 9 minutes (540 seconds), while the system solution development took 60 seconds, resulting in an efficiency improvement of 88.9%.

[0084] Scheme type Simple order processing time (minutes) Time taken for complex orders (in minutes) manual solution 8–10 15–25 System Solution 1 1

[0085] Table 3

[0086] Regarding order processing capacity, as shown in Table 4, the manual solution requires 20 people per person to handle 6 complex orders per day during peak periods (120 orders per day). This system, however, allows a single server to handle 30 orders concurrently per day, requiring only 3 operators for monitoring and assistance. For example, during the testing period, processing 100 orders on a certain day would require 17 people and 8 hours (a total of 136 person-hours); the system solution, using 3 people and 1 server, would take 8 hours (a total of 24 person-hours), resulting in an 82.4% reduction in manpower. The system processes each order in an average of 288 seconds.

[0087] Scheme type Daily order processing volume per person / machine Personnel required during peak periods manual solution 6 17 System Solution 30 3

[0088] Table 4

[0089] Regarding rework rate and cost, as shown in Table 3, rework refers to the process where workers, when implementing a stacking solution, need to unload products and re-retrieve them due to incorrect picking. The average rework rate for the manual solution is 12.5%. The test warehouse processes 12,000 orders annually, with each rework costing approximately 600 yuan (labor and time), resulting in an annual rework cost of approximately 900,000 yuan. The system solution reduces the rework rate to 3.5%, with an annual cost of approximately 252,000 yuan, representing a saving of 72.0%. For example, in 2024, a complex order (1,500 items, approximately 15 picks) required two reworks (costing 1,200 yuan) due to picking errors (14.2%), while the system solution (4.5% error rate) resulted in no rework and a cost of 0 yuan.

[0090] Scheme type Average rework rate (%) Annual rework cost (ten thousand yuan) manual solution 12.5 90.0 System Solution 3.5 25.2

[0091] Table 5

[0092] Therefore, based on the above characteristics, this invention overcomes the limitations of manual solutions in handling non-standard products. Compared to traditional two-dimensional drawings, three-dimensional models enable operators to understand the solutions more quickly, improving operational accuracy. It integrates the entire process of order input, data verification, grouping, stacking solution generation, and 3D visualization, drastically shortening process time and improving overall efficiency. This results in significant practical value and economic benefits.

[0093] A visualization modeling system for warehouse products implements the aforementioned visualization modeling method for warehouse products, including:

[0094] On the warehouse side, based on the information of the items, information is entered into the items and item data is generated;

[0095] The PC version is used to receive item data, model the items, and generate visual models.

[0096] The warehouse and PC terminals transmit data via a network.

[0097] The PC version includes an AI model that provides algorithms for modeling.

[0098] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A visual modeling method for warehouse products, characterized in that, Includes the following steps: S1. Obtain product order data; S2. Preprocess the order data to generate net data; S3. Construct a grouping processing module, which groups the products based on the net data to generate product packages; S4. Construct a stacking module. The stacking module performs stacking calculations on the products within the product package based on the net data and the product package, and generates a stacking scheme for the product package. S5. Construct a 3D modeling module, which performs 3D modeling of the product package based on the stacking scheme to generate a visual 3D model.

2. The visualization modeling method for warehouse products according to claim 1, characterized in that, The preprocessing includes data verification and data cleaning.

3. The visualization modeling method for warehouse products according to claim 1, characterized in that, The specific steps of the group processing module are as follows: S31. Obtain the size, weight, and shape of the product based on the net data; S32. Based on the size, weight, and shape, the products are grouped using a grouping algorithm to generate product packages.

4. The visualization modeling method for warehouse products according to claim 1, characterized in that, The specific operating steps of the stacking module are as follows: S41. Obtain the size, weight, and shape of the product based on the net data; S42. Based on the size, weight and shape, a stacking optimization algorithm is used to perform stacking calculations on the products in the product package to generate the stacking scheme.

5. The visualization modeling method for warehouse products according to claim 1, characterized in that, The content displayed in the visualized 3D model includes: The dimensions, weight, and total number of products of the visualized 3D model; the number and arrangement of products in each layer of the visualized 3D model; the process of stacking the products to form the visualized 3D model; the overall frame outline of the visualized 3D model in a semi-transparent state; and the different display effects of selected products in the visualized 3D model compared to unselected products.

6. A visual modeling system for warehouse products, characterized in that, Implementing the visual modeling method for warehouse products according to any one of claims 1-5, comprising: At the warehouse end, based on the information of the items, information is entered into the items and item data is generated; The PC terminal is used to receive the item data, model the item, and generate a visual model. The warehouse terminal and the PC terminal communicate via a network.

7. A visualization modeling system for warehouse products according to claim 6, characterized in that: The PC is equipped with an AI model, which provides algorithms and applies them to the modeling process.

Images