Big data-based method for matching a high-bay sorting rack

Abstract

The present application relates to the field of sorting rack matching, more particularly, the present application relates to a method for sorting rack matching based on big data. The method comprises: obtaining size information of each batch of order boards after the boards are cut, clustering the boards according to the size information to obtain a size clustering result; clustering the historical orders according to the size clustering result to obtain an order clustering result, obtaining the type of the sorting rack and the number of the sorting rack; matching the boards in the orders to be matched based on the type of the sorting rack and the number of the sorting rack, and completing the sorting rack matching when the degree of board mismatch is the smallest. Through the technical scheme of the present application, the accuracy of the sorting rack matching board result can be improved, thereby helping to reduce the labor cost.

Description

Technical Field

[0001] This invention relates to the field of sorting rack matching. More specifically, this invention relates to a big data-based method for matching sorting racks in automated warehouses. Background Technology

[0002] Automated storage and retrieval systems (AS / RS) are a key component of modern logistics and warehousing management, widely used in e-commerce, manufacturing, retail, and other sectors. Through automated sorting and storage technologies, these systems significantly improve logistics efficiency, reduce labor costs, and enhance overall operational flexibility and responsiveness. With the development of big data technology, massive amounts of logistics data can be collected and analyzed, providing new solutions for logistics management. In-depth analysis of logistics data can optimize inventory management, predict demand changes, and improve sorting efficiency, thereby enabling more intelligent and personalized logistics services. In AS / RS systems, achieving efficient shelf matching and goods allocation based on the characteristics of the goods and sorting requirements is a complex and challenging problem.

[0003] A Chinese patent application with publication number CN106485300A discloses a sorting method and system for sheet metal. The method performs image recognition on the sheet metal labels to obtain the sheet metal label information; it then performs a matching search in a component information database based on the sheet metal label information to obtain the component information to which the sheet metal belongs; the sheet metal is sorted through a sorting channel and stored in a warehouse, and the storage status of the sheet metal is monitored and recorded; if the monitoring record shows that there are already assembled sheet metal belonging to the same component in the warehouse, the assembled sheet metal is then combined and output through the sorting channel.

[0004] However, when there are unlabeled boards that need to be sorted, it will lead to re-labeling and matching, resulting in low matching efficiency for the boards. Summary of the Invention

[0005] To address the problem of low matching efficiency of sheet materials, this invention proposes a big data-based matching method for automated warehouse sorting racks.

[0006] This invention discloses a big data-based method for matching sorting racks in automated warehouses, comprising: acquiring the size information of sheet metal after cutting in each batch of orders in history; performing size clustering on the sheet metal based on the size information to obtain size clustering results; performing order clustering on historical orders based on the size clustering results to obtain order clustering results, thereby obtaining the type and quantity of sorting racks; matching the sheet metal in the orders to be matched based on the type and quantity of sorting racks; and completing the sorting rack matching when the sheet metal mismatch is minimized; the sheet metal mismatch satisfies the following relationship: ,in, Indicates the degree of mismatch between the boards. This indicates the number of different types of boards in the order. Indicates the first in the order The number of different types of boards, Indicates the first The number of sorting racks corresponding to each order cluster Indicates the first Each order cluster corresponds to the sorting rack storage. The quantity of different types of boards.

[0007] By using size clustering analysis, the boards in historical orders are classified according to their size specifications, thereby simplifying board inventory management. Further clustering of orders using the size clustering results not only helps identify groups of orders with similar needs, but also allows for the prediction of sorting rack type and quantity requirements based on order characteristics. This clustering method enables precise matching of orders and sorting rack resources, achieving efficient sorting and storage of boards.

[0008] Preferably, the step of size clustering includes: size information including length and width, calculating size cluster centers, and performing size clustering based on the length and width of the board and the size cluster centers; the formula for size clustering satisfies the following relation: ,in, Indicates board material and boards The distance between them and They represent the boards respectively Length and sheet material Length, and They represent the boards respectively Length and sheet material The width.

[0009] By calculating the cluster centers of sheet sizes, it is possible to determine the typical representative values ​​for each size category. These central values ​​summarize the main distribution characteristics of sheet sizes. This improves the efficiency of sheet storage and retrieval, and also helps optimize production planning and reduce material waste.

[0010] Preferably, the size cluster centers include: ,in, The length of the board material representing the size cluster center. Indicates the size of the cluster. The number of different types of boards, Indicates the size of the cluster. The length of the board material.

[0011] Similarly, obtain the board width of the size cluster center; use the calculated board length and board width as the size cluster center.

[0012] This method ensures that the size cluster centers accurately reflect the dimensional characteristics of the boards within their respective clusters. This not only simplifies the management of board size diversity but also provides a standardized reference for board classification, storage, and retrieval. Through size cluster centers, commonalities among various board types can be quickly identified, thereby optimizing inventory management and production planning, and improving material utilization and production efficiency.

[0013] Preferably, the order clustering includes: calculating order cluster centers, and clustering orders according to the size clustering results and the order cluster centers; the order clustering formula satisfies the following relation: ,in, Indicates order and orders The distance between them Indicates order The Middle The probability of similar boards appearing, Indicates order The Middle The probability of similar boards appearing.

[0014] This method measures the differences in the distribution of board categories among different orders, where the distance formula reflects the comparison between the probability of board category within an order and the probability of the order's cluster center. This approach effectively captures the similarities and differences between orders, allowing orders with similar board demand patterns to be clustered together.

[0015] Preferably, the order clustering formula further includes: treating orders as multi-dimensional vectors, with each dimension representing a type of board material; and using the similarity between two orders as the distance between the two orders.

[0016] Preferably, the order cluster centers satisfy the following relation: ,in, Indicates the first In the order cluster, the first Order clustering center for board-like materials Indicates the first Orders in an order cluster The Middle The number of different types of boards, This indicates the total number of orders.

[0017] Quickly identifying the types of sheet metal with concentrated demand enables more accurate decisions in inventory management. This helps improve production efficiency, reduce inventory backlog, speed up order response, and ultimately enhance customer satisfaction and the company's market competitiveness.

[0018] Preferably, obtaining the type and quantity of sorting racks includes: the type of sorting rack is equal to the distribution of order boards corresponding to the order cluster center; and the ratio of the quantity of each sorting rack is equal to the ratio of the number of orders in each order cluster.

[0019] Preferably, the process of completing the sorting rack matching further includes: matching a sorting rack with boards to obtain the best match; and re-matching the remaining boards in response to the fact that there are still unmatched boards in the order.

[0020] The beneficial effects of this invention are: 1. Through data analysis and clustering techniques, the storage and sorting process of sheet materials was deeply optimized. First, detailed size clustering was performed using the size information of sheet materials from historical orders, ensuring that the materials were classified in an orderly manner according to specifications. Then, the size clustering results were used to intelligently cluster orders, thereby accurately grasping the characteristics and needs of different orders. By calculating the sheet material mismatch, the type and quantity of sorting racks could be dynamically adjusted to achieve precise matching with order requirements, minimize mismatch, improve sheet material matching efficiency, and thus improve storage efficiency and retrieval speed.

[0021] 2. The automated matching process not only reduces manual operation and errors caused by human factors, but also significantly improves the efficiency and accuracy of sorting rack matching. This invention ensures that sorting rack configuration can be quickly adjusted to meet changes in production needs when faced with different types and quantities of orders. Real-time updated clustering results and matching degree calculations enable the system to respond quickly to real-time changes in orders, achieving instant matching and improving response speed and service quality. Attached Figure Description

[0022] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of the invention are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein: Figure 1 This is a flowchart of a big data-based matching method for automated warehouse sorting racks, according to an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] It should be understood that when the terms "first," "second," etc., are used in the claims, specification, and drawings of this invention, they are only used to distinguish different objects and not to describe a specific order. The terms "comprising" and "including" used in the specification and claims of this invention indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0025] This invention provides a big data-based method for matching sorting racks in automated warehouses. For example... Figure 1 As shown, the big data-based matching method for automated warehouse sorting racks includes steps S1-S3, which are described in detail below.

[0026] S1: Obtain the size information of the boards after cutting in each batch of orders in history, and perform size clustering on the boards based on the size information to obtain the size clustering results.

[0027] In one embodiment, accurately recording and analyzing the dimensional information of each batch of order sheets is crucial during the sheet processing and production management process. This information not only helps optimize the production process but also improves material utilization and product quality.

[0028] Specifically, we collect the dimensional information of the boards after cutting from each batch of orders in the historical data. The dimensional information includes length and width. Since the thickness of the boards is generally consistent, the thickness of the boards is not considered as dimensional information.

[0029] For irregularly shaped boards, the dimensions of their smallest bounding rectangle are used instead; that is, the length and width of this smallest bounding rectangle are measured and recorded.

[0030] Size clustering of the boards is performed based on their size information, including: selecting a preset number of initial center points, selecting a board, and calculating the distance between the board and each initial center point. The size clustering formula satisfies the following relationship: ,in, Indicates board material and boards The distance between them and They represent the boards respectively Length and sheet material Length, and They represent the boards respectively Length and sheet material The width.

[0031] The selected material is used to construct a size cluster with the initial center point that is closest to it.

[0032] Update the size cluster centers to satisfy the following relation: ,in, The length of the board material representing the size cluster center. Indicates the size of the cluster. The number of different types of boards, Indicates the size of the cluster. The length of the board material.

[0033] Similarly, obtain the board width of the size cluster center; use the calculated board length and board width as the size cluster center.

[0034] Next, select a new board material and calculate its distance to the three center points, which include an updated size cluster center and two initial center points. Iterate through all boards to complete size clustering. After clustering, use the profile coefficient method to select the best clustering result as the size clustering result.

[0035] In one embodiment, the K-means clustering algorithm is used to cluster the boards.

[0036] S2, based on the size clustering results, perform order clustering on historical orders to obtain the order clustering results, and obtain the type and quantity of sorting racks.

[0037] In one embodiment, historical orders are clustered based on the size clustering results obtained in step S1 to determine the quantity of different types of boards in the orders, thereby obtaining the type and quantity of sorting racks.

[0038] Specifically, based on the clustering process in step S1, the orders are clustered, and the formula for order clustering satisfies the following relation: ,in, Indicates order and orders The distance between them Indicates order The Middle The probability of similar boards appearing, Indicates order The Middle The probability of similar boards appearing.

[0039] Calculate the order cluster centers, which satisfy the following relation: ,in, Indicates the first In the order cluster, the first Order clustering center for board-like materials Indicates the first Orders in an order cluster The Middle The number of different types of boards, This indicates the total number of orders.

[0040] At this point, we can obtain the order clustering results.

[0041] The type of sorting rack is equal to the distribution of order boards corresponding to the order cluster center; the ratio of the number of sorting racks is equal to the ratio of the number of orders in each order cluster.

[0042] For example, the board sizes are divided into extra-large, large, medium, small, and extra-small boards. The order cluster center corresponds to three types of board sizes: large, small, and extra-small. Therefore, there are also three types of sorting racks. When clustering orders, three order clusters are obtained, with 50, 40, and 70 orders in each cluster, respectively. Therefore, the ratio of sorting racks should be 5:4:7.

[0043] In one embodiment, the formula for order clustering further includes: treating orders as multi-dimensional vectors, where each dimension represents the quantity of a type of board; and using the similarity between two orders as the distance between the two orders.

[0044] S3, based on the type and quantity of sorting racks, matches the boards in the orders that need to be matched. When the board mismatch is minimal, the sorting rack matching is completed.

[0045] In one embodiment, the size information of the boards in the orders to be matched is obtained, and the size cluster to which each board belongs is calculated. The distribution of boards in the orders to be matched is obtained, and the optimal combination of sorting racks is determined based on the distribution of boards in the orders to be matched and the number of boards that different sorting racks can store. The board mismatch degree is calculated, and the board mismatch degree satisfies the following relationship: ,in, Indicates the degree of mismatch between the boards. This indicates the number of different types of boards in the order. Indicates the first in the order The number of different types of boards, Indicates the first The number of sorting racks corresponding to each order cluster Indicates the first Each order cluster corresponds to the sorting rack storage. The quantity of different types of boards.

[0046] When the mismatch between the boards is minimized, the optimal sorting rack combination is obtained, and sorting rack matching is completed.

[0047] In one embodiment, a sorting rack is matched with boards to obtain the best match; in response to the fact that there are still unmatched boards in the order, the remaining boards are rematched.

[0048] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

[0049] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A big data-based matching method for automated warehouse sorting racks, characterized in that, include: Obtain the dimensional information of the boards after cutting from each batch of orders in history, and perform dimensional clustering on the boards based on the dimensional information to obtain the dimensional clustering results; The size clustering includes: The size information includes length and width. The size cluster centers are calculated, and size clustering is performed based on the length and width of the board and the size cluster centers. Select a preset number of initial center points, select a board material, and calculate the distance between the board material and each initial center point; The formula for size clustering satisfies the following relation: ,in, Indicates board material and boards The distance between them and They represent the boards respectively Length and sheet material Length, and They represent the boards respectively Width and sheet material The width; The selected board material is used to construct a size cluster with the initial center point that has the shortest distance; The size cluster centers include: ,in, The length of the board material representing the size cluster center. Indicates the size of the cluster. The number of different types of boards, Indicates the size of the cluster. The length of the board material; Similarly, the width of the plate at the size cluster center is obtained; The calculated board length and board width are used as the size cluster centers; Based on the size clustering results, historical orders are clustered to obtain the order clustering results, which in turn determine the type and quantity of sorting racks, including: The type of sorting rack is equal to the distribution of order boards corresponding to the order cluster center; The ratio of the number of sorting racks is equal to the ratio of the number of orders in each order cluster; Based on the type and quantity of sorting racks, the boards in the orders to be matched are matched. When the mismatch of the boards is minimized, the sorting rack matching is completed. The mismatch between the sheet materials satisfies the following relationship: ;in Indicates the degree of mismatch between the boards. This indicates the number of different types of boards in the order. Indicates the first in the order The number of different types of boards, Indicates the first The number of sorting racks corresponding to each order cluster Indicates the first Each order cluster corresponds to the sorting rack storage. The quantity of different types of boards; The order clustering process includes: Calculate order cluster centers, and cluster orders based on the size clustering results and order cluster centers; The formula for order clustering satisfies the following relation: ,in, Indicates order and orders The distance between them Indicates order The Middle The probability of similar boards appearing, Indicates order The Middle The probability of similar boards appearing; The order cluster centers satisfy the following relation: ,in, Indicates the first In the order cluster, the first Order clustering center for board-like materials Indicates the first Orders in an order cluster The Middle The number of different types of boards, This indicates the total number of orders.

2. The big data-based matching method for automated warehouse sorting racks according to claim 1, characterized in that, The formula for order clustering also includes: The orders are treated as a multi-dimensional vector, with each dimension representing a type of board material. The similarity between two orders is used as the distance between the two orders.

3. The big data-based matching method for automated warehouse sorting racks according to claim 1, characterized in that, The type and number of sorting racks obtained include: The type of sorting rack is equal to the distribution of order boards corresponding to the order cluster center; The ratio of the number of sorting racks is equal to the ratio of the number of orders in each order cluster.

4. The big data-based matching method for automated warehouse sorting racks according to claim 1, characterized in that, The completion of sorting rack matching also includes: Match a sorting rack with the appropriate sheet material to achieve the best match; If there are still unmatched boards in the order, the remaining boards will be rematched.

Images