Warehouse picking path optimization method and system based on self-defined loop point clustering

Abstract

The application relates to a warehouse picking path optimization method based on a self-defined loop point clustering, which comprises the following steps: importing order data and carrying out pretreatment, wherein the order data comprises an order number and a storage location code; presetting the order quantity contained in each picking order according to the order data, and determining a distance matrix according to the order number and the storage location code; determining an optimal picking path through a self-defined loop point clustering algorithm according to the distance matrix, wherein the optimal picking path is the sum of all picking order total paths; and generating a picking order according to the optimal picking path. Through the picking route optimized by the application, orders of adjacent storage locations are distributed together as much as possible to carry out picking, the shelves required by each picking personnel are concentrated as much as possible, all goods contained in the picking order are sequentially searched according to the cabinet sequence and classified according to the order number, the picking personnel only needs to walk once, the total picking path of all picking personnel is reduced as much as possible, and the overall efficiency of warehouse picking is improved.

Description

Technical Field

[0001] This invention relates to the field of warehousing and logistics technology, and in particular to a warehouse picking route optimization method and system based on custom point clustering. Background Technology

[0002] In e-commerce logistics warehouses, goods are distributed across different shelves. When a consumer places an order on an e-commerce platform, the order flows into the warehouse system, where warehouse staff begin picking the goods. As the core of order processing, picking is the most time-consuming and labor-intensive operation in the warehouse, and the walking distance directly impacts the picking efficiency of the pickers. When the number of orders is large, the warehouse will combine multiple orders into a single picking list for picking. However, traditional warehouse picking methods typically involve finding all the items for each order before processing the next order. This causes pickers to repeatedly traverse the same shelves, increasing the total picking path and resulting in low picking efficiency.

[0003] Therefore, there is an urgent need to propose a warehouse picking route optimization method to solve the aforementioned technical problems. Summary of the Invention

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the problems of long total picking distance and low picking efficiency in the prior art, and to propose a warehouse picking path optimization method and system based on custom point clustering.

[0005] To address the aforementioned technical problems, this invention provides a warehouse picking route optimization method based on custom point clustering, comprising the following steps:

[0006] S1. Import order data and preprocess it, wherein the order data includes order number and storage location code;

[0007] S2. Preset the number of orders included in each picking list based on the order data, and determine the distance matrix based on the order number and the storage location code;

[0008] S3. Based on the distance matrix, find cluster points using a custom point-following clustering algorithm, and determine whether the number of cluster points reaches the order quantity of the picking list. If the determination is yes, generate the picking path of a single picking list through the clusters completed by each cluster. If the determination result is no, replenish the cluster points to the order quantity of the picking list at once, and generate the picking path of a single picking list through the clusters completed by each cluster.

[0009] S4. Calculate the optimal picking path based on all picking paths, and generate picking lists based on the optimal picking path. The optimal picking path is the one with the smallest sum of the total paths of all picking lists.

[0010] In one embodiment of the present invention, the order data is imported in S1 by reading a file containing order numbers and storage location codes. Each order number in the order data contains at least one item, each item corresponds to a storage container number, and the storage location code represents the location information of the item.

[0011] In one embodiment of the present invention, the method for preprocessing order data in S1 includes:

[0012] The picking distance for each picker is the maximum location code minus the minimum location code.

[0013] For each order number, only the maximum and minimum values ​​of the corresponding location codes for multiple items are retained. If an order has only one item, its maximum and minimum values ​​are the same.

[0014] In one embodiment of the present invention, the method for presetting the number of orders included in each picking list based on the order data in step S2 includes:

[0015] The number of orders contained in each picking list is preset. Each order data generates m picking lists, and each picking list contains n orders.

[0016] In one embodiment of the present invention, the method for determining the distance matrix based on the order number and the storage location code in step S2 includes:

[0017] The distance matrix includes the order number, the maximum location code, and the minimum location code. The distance matrix is ​​represented as a two-dimensional coordinate graph with the maximum and minimum location codes as coordinates. The picking path of each picking list is the value of the maximum location code minus the minimum location code among all the location codes of the n orders it contains.

[0018] In one embodiment of the present invention, the method for generating picking paths using a custom point-following clustering algorithm in step S3 includes:

[0019] S31. The total path of each picking list is the value of the largest location code minus the smallest location code among all the location codes of the n orders it contains;

[0020] S32. The custom point-based clustering algorithm uses Euclidean distance to find the nearest point to the lower right of the previously found point and then repeats the above steps based on the newly found points until n points are found.

[0021] S33. Use a custom point-following clustering algorithm to find all clusters that meet the conditions. The characteristic of a cluster that meets the conditions is that the first point is the starting point and the remaining n-1 points are all to the lower right of the starting point. If there are no points to the lower right of a certain point during the point-following clustering process, that is, if n points are not found, then find the nearest points around the last found point according to the Euclidean distance and sort them to fill the n points at a time.

[0022] S34. For each cluster, the difference between the maximum and minimum location codes among all points in the cluster is the picking path of a single picking order. All points contained in the cluster are all the order numbers of the corresponding picking order.

[0023] S35. The sum of picking paths for all picking orders is the total picking path. The optimal picking path is determined based on the total picking path.

[0024] In one embodiment of the present invention, the Euclidean distance in S32 is Where x1 and y1 are the minimum and maximum storage location codes for the first order number, respectively, and x2 and y2 are the minimum and maximum storage location codes for the second order number, respectively, and they always cluster to the lower right of the starting point.

[0025] Furthermore, this invention also provides a warehouse picking route optimization system based on custom point clustering, comprising:

[0026] The data processing module is used to import order data and perform preprocessing, wherein the order data includes order number and storage location code;

[0027] The distance matrix determination module presets the number of orders included in each picking list based on the order data. The distance matrix determination module is used to determine the distance matrix based on the order number and the storage location code.

[0028] A custom point-following clustering module is used to find cluster points based on a distance matrix using a custom point-following clustering algorithm. It determines whether the number of cluster points reaches the order quantity of the picking list. If the determination is yes, the picking path of a single picking list is generated through the clusters completed by each cluster. If the determination result is no, the number of cluster points to reach the order quantity of the picking list is replenished at once, and the picking path of a single picking list is generated through the clusters completed by each cluster.

[0029] The picking list generation module is used to calculate the optimal picking path based on all picking paths and generate picking lists based on the optimal picking path, wherein the optimal picking path is the one with the smallest sum of the total paths of all picking lists.

[0030] In one embodiment of the present invention, the method by which the distance matrix determination module determines the distance matrix based on the order number and the storage location code includes:

[0031] The distance matrix includes the order number, the maximum location code, and the minimum location code. The distance matrix is ​​represented as a two-dimensional coordinate graph with the maximum and minimum location codes as coordinates. The picking path of each picking list is the value of the maximum location code minus the minimum location code among all the location codes of the n orders it contains.

[0032] In one embodiment of the present invention, the method for generating picking paths using a custom point-following clustering module and a custom point-following clustering algorithm includes:

[0033] The total path of each picking list is the value of the largest location code minus the smallest location code among all the location codes of the n orders it contains;

[0034] The custom point-based clustering algorithm uses Euclidean distance to find the nearest point to the lower right of the previously found point and clusters them together. Then, it repeats the above steps based on the newly found points until n points are found.

[0035] A custom point-following clustering algorithm is used to find all clusters that meet the conditions. The characteristic of a cluster that meets the conditions is that the first point is the starting point, and the remaining n-1 points are all to the lower right of the starting point. If there are no points to the lower right of a certain point during the point-following clustering process, that is, if n points are not found, the nearest points are found around the last found point according to the Euclidean distance and sorted, and the n points are filled in one go.

[0036] The difference between the maximum and minimum location codes among all points in the cluster is the picking path of a single picking order. All points contained in the cluster are all the order numbers of the corresponding picking order.

[0037] The sum of picking paths for all picking orders is the total picking path. The optimal picking path is determined based on the total picking path.

[0038] The technical solution of the present invention has the following advantages compared with the prior art:

[0039] This invention optimizes the picking route to group orders with adjacent storage locations together for picking, ensuring that the shelves that each picker needs to pick are concentrated as much as possible. All items in the picking list are searched sequentially according to the shelf order and classified by order number. Pickers only need to walk once, minimizing the total picking path for all pickers and improving the overall efficiency of warehouse picking. Attached Figure Description

[0040] Figure 1This is a flowchart of a warehouse picking path optimization method based on custom point clustering proposed in an embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram illustrating the preprocessing of the first nine items of order data as an example in the method of the present invention.

[0042] Figure 3 The method of this invention represents the distance matrix as a two-dimensional coordinate graph with the maximum and minimum cargo location codes as coordinates.

[0043] Figure 4 This is a schematic diagram of the clusters found using the custom point-following clustering algorithm of this invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0045] Reference Figure 1 As shown, this invention discloses a warehouse picking route optimization method based on custom point clustering, comprising the following steps:

[0046] S1. Import order data and preprocess it, wherein the order data includes order number and storage location code;

[0047] S2. Preset the number of orders included in each picking list based on the order data, and determine the distance matrix based on the order number and the storage location code;

[0048] S3. Based on the distance matrix, find cluster points using a custom point-following clustering algorithm, and determine whether the number of cluster points reaches the order quantity of the picking list. If the determination is yes, generate the picking path of a single picking list through the clusters completed by each cluster. If the determination result is no, replenish the cluster points to the order quantity of the picking list at once, and generate the picking path of a single picking list through the clusters completed by each cluster.

[0049] S4. Calculate the optimal picking path based on all picking paths, and generate picking lists based on the optimal picking path. The optimal picking path is the one with the smallest sum of the total paths of all picking lists.

[0050] In a warehouse picking path optimization method based on custom point clustering disclosed in this invention embodiment, step S1 imports order data by reading an Excel spreadsheet containing order numbers and location codes. Each order number contains at least one item, each item corresponds to a storage container number, and the location code represents the item's location information. The picking path distance for each picker is the maximum location code minus the minimum location code among all the locations they need to traverse. For each order number, only the maximum and minimum values ​​of the corresponding location codes for multiple items are retained. If an order contains only one item, its maximum and minimum values ​​are the same. Figure 2 The first 9 items of imported order data were selected for data preprocessing.

[0051] In a warehouse picking path optimization method based on custom point clustering disclosed in an embodiment of the present invention, S2 presets the number of orders contained in each picking order, wherein each order data generates m picking orders, and each picking order contains n orders.

[0052] In a warehouse picking path optimization method based on custom point clustering disclosed in an embodiment of the present invention, the distance matrix in S2 consists of the order number, the maximum storage location code, and the minimum storage location code, and is represented as a two-dimensional coordinate graph with the maximum and minimum storage location codes as coordinates. Figure 3 In the two-dimensional coordinate graph shown, each point represents the coordinates of the minimum and maximum storage location codes for the corresponding order number. The coordinates of the i-th order number are represented as (x... i ,y i ), where x i y represents the value of the minimum location code for the i-th order number. i This represents the value of the maximum location code for the i-th order number. It should be noted that when multiple orders have the same maximum and minimum location codes, their coordinates will overlap. The picking path for each picking list is the value of the maximum location code minus the minimum location code among all the location codes of its n contained orders.

[0053] In a warehouse picking route optimization method based on custom point clustering disclosed in an embodiment of the present invention, the method in S3 for determining the optimal picking route based on the distance matrix using a custom point clustering algorithm includes:

[0054] S31. The total path of each picking list is: the value of the largest location code minus the smallest location code among all the location codes of the n orders it contains;

[0055] S32. Using a custom point-based clustering algorithm, which employs Euclidean distance, the algorithm finds the nearest points to the lower right of the previously found point and clusters them together. This process is repeated based on newly found points until n points have been found. The Euclidean distance is used... Where x1 and y1 are the minimum and maximum storage location codes for the first order number, respectively, and x2 and y2 are the minimum and maximum storage location codes for the second order number, respectively. For example... Figure 4 It always clusters to the lower right of the starting point, so that the coordinates of the starting point represent the minimum and maximum location codes of its picking list. Each cluster only needs to calculate the difference between the coordinates of the starting point. Compared with traditional dynamic path planning, it does not need to repeatedly calculate the other points in the cluster, thus achieving a balance between the time consumed in generating the path and the path length.

[0056] S33. Use the above algorithm to find all clusters that meet the conditions. The characteristic of a cluster that meets the conditions should be that the first point is the starting point and the remaining n-1 points are all to the lower right of the starting point.

[0057] S34. If there are no points to the right of a certain point that meet the conditions during the point clustering process, that is, if n points are not found, then find the nearest points around the last found point according to the Euclidean distance and sort them to fill the n points at once.

[0058] S35. The clusters completed by each cluster contain all the points that are the order numbers of the corresponding picking list.

[0059] S36. The picking path of a single picking list is the difference between the maximum and minimum location codes among all point coordinates in the cluster.

[0060] S37. The total picking path is the sum of the picking paths of all picking orders.

[0061] In a warehouse picking path optimization method based on custom point clustering disclosed in an embodiment of the present invention, in step S4, the m clusters obtained from the clustering results in step S35 are matched with the order number and location code in the original order data respectively to generate m picking orders containing the order number and location code.

[0062] The following describes a warehouse picking path optimization system based on custom point clustering disclosed in an embodiment of the present invention. The warehouse picking path optimization system based on custom point clustering described below and the warehouse picking path optimization method based on custom point clustering described above can be referred to and correspond to each other.

[0063] This invention also provides a warehouse picking route optimization system based on custom point clustering, comprising:

[0064] The data processing module is used to import order data and perform preprocessing, wherein the order data includes order number and storage location code;

[0065] The distance matrix determination module presets the number of orders included in each picking list based on the order data. The distance matrix determination module is used to determine the distance matrix based on the order number and the storage location code.

[0066] A custom point-following clustering module is used to find cluster points based on a distance matrix using a custom point-following clustering algorithm. It determines whether the number of cluster points reaches the order quantity of the picking list. If the determination is yes, the picking path of a single picking list is generated through the clusters completed by each cluster. If the determination result is no, the number of cluster points to reach the order quantity of the picking list is replenished at once, and the picking path of a single picking list is generated through the clusters completed by each cluster.

[0067] The picking list generation module is used to calculate the optimal picking path based on all picking paths and generate picking lists based on the optimal picking path, wherein the optimal picking path is the one with the smallest sum of the total paths of all picking lists.

[0068] In one embodiment of the present invention, the method by which the distance matrix determination module determines the distance matrix based on the order number and the storage location code includes:

[0069] The distance matrix includes the order number, the maximum location code, and the minimum location code. The distance matrix is ​​represented as a two-dimensional coordinate graph with the maximum and minimum location codes as coordinates. The picking path of each picking list is the value of the maximum location code minus the minimum location code among all the location codes of the n orders it contains.

[0070] In one embodiment of the present invention, the method for generating picking paths using a custom point-following clustering module and a custom point-following clustering algorithm includes:

[0071] The total path of each picking list is the value of the largest location code minus the smallest location code among all the location codes of the n orders it contains;

[0072] The custom point-based clustering algorithm uses Euclidean distance to find the nearest point to the lower right of the previously found point and clusters them together. Then, it repeats the above steps based on the newly found points until n points are found.

[0073] A custom point-following clustering algorithm is used to find all clusters that meet the conditions. The characteristic of a cluster that meets the conditions is that the first point is the starting point, and the remaining n-1 points are all to the lower right of the starting point. If there are no points to the lower right of a certain point during the point-following clustering process, that is, if n points are not found, the nearest points are found around the last found point according to the Euclidean distance and sorted, and the n points are filled in one go.

[0074] The difference between the maximum and minimum location codes among all points in the cluster is the picking path of a single picking order. All points contained in the cluster are all the order numbers of the corresponding picking order.

[0075] The sum of picking paths for all picking orders is the total picking path. The optimal picking path is determined based on the total picking path.

[0076] The warehouse picking route optimization system based on custom point clustering in this embodiment is used to implement the aforementioned warehouse picking route optimization method based on custom point clustering. Therefore, the specific implementation of this system can be found in the embodiment section of the warehouse picking route optimization method based on custom point clustering mentioned above. Thus, its specific implementation can be referred to the description of the corresponding embodiments, and will not be elaborated here.

[0077] Furthermore, since the warehouse picking path optimization system based on custom point clustering in this embodiment is used to implement the aforementioned warehouse picking path optimization method based on custom point clustering, its function corresponds to the function of the above method, and will not be repeated here.

[0078] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0080] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0082] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A warehouse picking route optimization method based on custom point clustering, characterized in that, Includes the following steps: S1. Import order data and preprocess it, wherein the order data includes order number and storage location code; S2. Preset the number of orders included in each picking list based on the order data, and determine the distance matrix based on the order number and the location code; S3. Based on the distance matrix, find cluster points using a custom point-following clustering algorithm, and determine whether the number of cluster points reaches the order quantity of the picking list. If the determination is yes, generate the picking path of a single picking list through the clusters completed by each cluster. If the determination result is no, replenish the cluster points to the order quantity of the picking list at once, and generate the picking path of a single picking list through the clusters completed by each cluster. S4. Calculate the optimal picking route based on all picking routes, and generate picking lists based on the optimal picking route. The optimal picking route is the one with the smallest sum of the total routes of all picking lists. The method for generating picking routes using a custom point-based clustering algorithm in S3 includes: S31. The total path of each picking list is the value of the largest location code minus the smallest location code among all the location codes of the orders it contains; S32. The custom point-based clustering algorithm uses Euclidean distance to find the nearest point to the lower right of the previously found point and then repeats the above steps based on the newly found points until all points have been found. S33. Use a custom point-following clustering algorithm to find all clusters that meet the conditions. The characteristic of a cluster that meets the conditions is that the first point is the starting point and the remaining points are all to the lower right of the starting point. If there are no points to the lower right of a certain point that meet the conditions during the point-following clustering process, that is, if not all points are found, then find the nearest points around the last found point according to the Euclidean distance and sort them, and fill in n points at a time. S34. For each cluster, the difference between the maximum and minimum location codes among all points in the cluster is the picking path of a single picking order. All points contained in the cluster are all the order numbers of the corresponding picking order. S35. The sum of the picking paths of all picking orders is the total picking path.

2. The warehouse picking route optimization method based on custom point clustering according to claim 1, characterized in that, The order data is imported in S1 by reading a file containing order numbers and storage location codes. Each order number in the order data contains at least one item, each item corresponds to a storage container number, and the storage location code indicates the location information of the item.

3. The warehouse picking route optimization method based on custom point clustering according to claim 1, characterized in that, The method for preprocessing order data in S1 includes: The picking distance for each picker is the maximum location code minus the minimum location code. For each order number, only the maximum and minimum values ​​of the corresponding location codes among multiple items are retained. If an order has only one item, its maximum and minimum values ​​are the same.

4. The warehouse picking route optimization method based on custom point clustering according to claim 3, characterized in that, The method in S2 for pre-setting the number of orders included in each picking list based on the order data includes: The number of orders contained in each picking list is preset. Each order data generates a picking list, and each picking list contains orders.

5. The warehouse picking route optimization method based on custom point clustering according to claim 4, characterized in that, The method for determining the distance matrix based on the order number and storage location code in S2 includes: The distance matrix includes the order number, the maximum location code, and the minimum location code. The distance matrix is ​​represented as a two-dimensional coordinate graph with the maximum and minimum location codes as coordinates. The picking path of each picking order is the value of the maximum location code minus the minimum location code among all the location codes of the orders it contains.

6. The warehouse picking route optimization method based on custom point clustering according to claim 1, characterized in that, In S32, the Euclidean distance is given by , where and are the minimum and maximum storage location codes of the first order number, respectively, and and are the minimum and maximum storage location codes of the second order number, respectively, and they always cluster to the lower right of the starting point.

7. A warehouse picking route optimization system based on custom point clustering, characterized in that, include: The data processing module is used to import order data and perform preprocessing, wherein the order data includes order number and storage location code; The distance matrix determination module presets the number of orders included in each picking list based on the order data. The distance matrix determination module is used to determine the distance matrix based on the order number and the storage location code. A custom point-following clustering module is used to find cluster points based on a distance matrix using a custom point-following clustering algorithm. It determines whether the number of cluster points reaches the order quantity of the picking list. If the determination is yes, the picking path of a single picking list is generated through the clusters completed by each cluster. If the determination result is no, the number of cluster points to reach the order quantity of the picking list is replenished at once, and the picking path of a single picking list is generated through the clusters completed by each cluster. The picking route is generated using a custom point-based clustering algorithm, which includes: the total route of each picking order is the value of the largest location code minus the smallest location code among all the location codes of the orders it contains; The custom point-based clustering algorithm uses Euclidean distance to find the nearest point to the lower right of the previously found point to form a cluster. This process is repeated with newly found points until all n points are found. The algorithm finds all clusters that meet certain conditions: the first point is the starting point, and all other points are to the lower right of the starting point. If no point meets the conditions to the lower right of a given point during the clustering process (i.e., not all n points are found), the algorithm searches for the nearest points around the last found point using Euclidean distance, sorts them, and fills the remaining n points at a time. For each completed cluster, the difference between the maximum and minimum location codes among all points in the cluster represents the picking path for a single picking order. All points within the cluster represent all order numbers for the corresponding picking order. The sum of all picking paths for all picking orders is the total picking path. The picking list generation module is used to calculate the optimal picking path based on all picking paths and generate picking lists based on the optimal picking path, wherein the optimal picking path is the one with the smallest sum of the total paths of all picking lists.

8. The warehouse picking route optimization system based on custom point clustering according to claim 7, characterized in that, The distance matrix determination module determines the distance matrix based on the order number and storage location code using the following methods: The distance matrix includes the order number, the maximum location code, and the minimum location code. The distance matrix is ​​represented as a two-dimensional coordinate graph with the maximum and minimum location codes as coordinates. The picking path of each picking order is the value of the maximum location code minus the minimum location code among all the location codes of the orders it contains.

9. The warehouse picking route optimization system based on custom point clustering according to claim 8, characterized in that, The method for generating picking routes using the custom point-following clustering module and the custom point-following clustering algorithm includes: The total path of each picking list is the value of the largest location code minus the smallest location code among all the location codes of the orders it contains; The custom point-based clustering algorithm uses Euclidean distance to find the nearest point to the lower right of the previously found point and then repeats the above steps based on the newly found points until all points have been found. A custom point-following clustering algorithm is used to find all clusters that meet the conditions. The characteristic of a cluster that meets the conditions is that the first point is the starting point and the remaining points are all to the lower right of the starting point. If there are no points to the lower right of a point that meet the conditions during the point-following clustering process, that is, if not all points are found, then the nearest points around the last found point are found according to the Euclidean distance and sorted, and n points are filled at a time. The difference between the maximum and minimum location codes among all points in the cluster is the picking path of a single picking order. All points contained in the cluster are all the order numbers of the corresponding picking order. The sum of picking paths for all picking orders is the total picking path. The optimal picking path is determined based on the total picking path.

Images