Storage location determination method and device, electronic equipment and computer readable medium

By receiving storage requests and historical data of items to be put on, cluster analysis and dimensional score calculations, and target storage locations are solved, and the storage locations are unreasonable and low outbound efficiency caused by the storage in the existing technology is achieved. More efficient item processing and storage space selection.

CN119090392APending Publication Date: 2024-12-06BEIJING JINGDONG QIANSHITECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310656224.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-06-05
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing methods of inventory entry and listing are only carried out in the order of inventory entry or customary methods, resulting in unreasonable storage locations and low out-of-stock efficiency.

Method used

By receiving the storage request for items to be put on, obtaining its attribute data and historical data of candidate storage positions, performing cluster analysis to determine the storage type, extracting dimensional feature data based on the preset dimension, calculating dimension scores, and determining the target storage position.

Benefits of technology

It improves the efficiency of item processing and accuracy of storage location selection, ensuring that items can be shipped out of the warehouse faster and more accurately.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090392A_ABST
    Figure CN119090392A_ABST
Patent Text Reader

Abstract

The invention discloses a storage location determination method and device, electronic equipment and a computer readable medium, and relates to the technical field of computers, one specific embodiment comprises the steps of receiving a storage request of a to-be-shelved article, and obtaining attribute data of the corresponding to-be-shelved article and historical data of candidate storage locations; performing clustering analysis on the historical data to determine a corresponding storage location type; based on a preset dimension, extracting corresponding dimension feature data in the attribute data of the to-be-put-on-shelf article, and determining a corresponding dimension score according to the dimension feature data; and according to the dimension score and the storage location type, determining a target storage location in the candidate storage locations and outputting the target storage location. And the article processing efficiency and the storage position selection accuracy are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a storage location determination method, device, electronic device, and computer-readable medium. Background Art

[0002] With the rapid development of e-commerce and modern logistics, more and more users are buying items online. After a wide range of items are ordered in the order system, the order system will send the order-related item information to the warehouse management system for positioning, pre-occupancy of inventory, and generation of waves to facilitate the subsequent picking and packing process. The current existing warehousing and shelving methods only enter the warehouse in the order of the warehouse entry order or in the most familiar way of entering the warehouse, resulting in unreasonable storage locations of items and the inability to exit the warehouse faster during the outbound operation, resulting in low efficiency in item processing. Summary of the invention

[0003] In view of this, the embodiments of the present application provide a storage location determination method, device, electronic device and computer-readable medium, which can solve the problem that the existing warehousing and shelving method only carries out warehousing in the order of the warehousing order or in the most common warehousing method, resulting in unreasonable storage location of items and the inability to be out of the warehouse faster during the outbound operation, resulting in low efficiency in item processing.

[0004] To achieve the above object, according to one aspect of an embodiment of the present application, a storage location determination method is provided, comprising:

[0005] Receive a storage request for items to be put on the shelves, and obtain attribute data of the items to be put on the shelves and historical data of candidate storage locations;

[0006] Perform cluster analysis on historical data to determine the corresponding storage type;

[0007] Extracting corresponding dimension feature data from the attribute data of the items to be put on the shelves based on the preset dimensions, and determining corresponding dimension scores according to the dimension feature data;

[0008] According to the dimension score and storage location type, the target storage location among the candidate storage locations is determined and output.

[0009] Optionally, cluster analysis is performed on the historical data to determine the corresponding storage location type, including:

[0010] Extract corresponding storage location feature data from historical data based on preset dimensions;

[0011] The storage location feature data is input into the cluster analysis model to output the corresponding storage location type, wherein the cluster analysis model is determined by the following steps:

[0012] Obtaining sample feature data of a preset dimension and a storage location type corresponding to the sample feature data;

[0013] The sample feature data is used as the input of the initial cluster analysis model, and the storage location type corresponding to the sample feature data is used as the expected output of the initial cluster analysis model to train the initial cluster analysis model and generate a final cluster analysis model.

[0014] Optionally, determining a corresponding dimension score includes:

[0015] The dimension feature data is input into the storage location classification selection model, and the dimension score of the corresponding preset dimension is obtained through the decision tree in the storage location classification selection model.

[0016] Optionally, determining a target storage location among candidate storage locations according to the dimension score and the storage location type includes:

[0017] Summarize the dimension scores of the preset dimensions according to the storage location type to determine the dimension score corresponding to each storage location type;

[0018] Calculate the mean of the dimension scores corresponding to each storage location type;

[0019] According to the mean, the target storage location type is determined, and then the target storage location among the candidate storage locations is determined according to the target storage location type.

[0020] Optionally, determining a target storage location among candidate storage locations according to the target storage location type includes:

[0021] Determine a storage location among the candidate storage locations that corresponds to the target storage location type;

[0022] The configuration data of the storage location corresponding to the target storage location type is obtained, and the target storage location is determined according to the attribute data and configuration data of the items to be put on the shelves.

[0023] Optionally, before inputting the dimensional feature data into the storage location classification selection model, the method further includes:

[0024] Obtain a sample set, wherein the sample set includes feature data of preset dimensions of historical items on the shelves, feature data of preset dimensions of historical items off the shelves, and corresponding sample storage location types;

[0025] Model training is performed based on the feature data of preset dimensions of historical items on the shelves, the feature data of preset dimensions of historical items off the shelves, and the corresponding sample storage location types to obtain a storage location classification selection model.

[0026] Optionally, a storage location classification selection model is obtained, including:

[0027] Generate a corresponding decision tree based on the input feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves;

[0028] The generated decision trees are used to form a storage location classification selection model.

[0029] Optionally, a corresponding decision tree is generated, including:

[0030] Determine the information entropy corresponding to the feature data of the preset dimension of the historical listing items and the feature data of the preset dimension of the historical delisting items;

[0031] Target feature data is obtained by screening according to information entropy, and the target feature data is combined based on preset dimensions to obtain combined target feature data;

[0032] Generate a corresponding decision tree based on the combined target feature data.

[0033] In addition, the present application also provides a storage position determination device, comprising:

[0034] A receiving unit is configured to receive a storage request for items to be put on the shelves, and obtain attribute data of the items to be put on the shelves and historical data of candidate storage locations;

[0035] A cluster analysis unit, configured to perform cluster analysis on the historical data to determine a corresponding storage location type;

[0036] A dimension score determination unit is configured to extract corresponding dimension feature data from the attribute data of the items to be put on the shelf based on the preset dimension, and determine the corresponding dimension score according to the dimension feature data;

[0037] The target storage location determination unit is configured to determine and output a target storage location among candidate storage locations based on the dimension score and the storage location type.

[0038] Optionally, the cluster analysis unit is further configured to:

[0039] Extract corresponding storage location feature data from historical data based on preset dimensions;

[0040] The storage location feature data is input into the cluster analysis model to output the corresponding storage location type, wherein the cluster analysis model is determined by the following steps:

[0041] Obtaining sample feature data of a preset dimension and a storage location type corresponding to the sample feature data;

[0042] The sample feature data is used as the input of the initial cluster analysis model, and the storage location type corresponding to the sample feature data is used as the expected output of the initial cluster analysis model to train the initial cluster analysis model and generate a final cluster analysis model.

[0043] Optionally, the dimension score determination unit is further configured to:

[0044] The dimension feature data is input into the storage location classification selection model, and the dimension score of the corresponding preset dimension is obtained through the decision tree in the storage location classification selection model.

[0045] Optionally, the target storage location determination unit is further configured to:

[0046] Summarize the dimension scores of the preset dimensions according to the storage location type to determine the dimension score corresponding to each storage location type;

[0047] Calculate the mean of the dimension scores corresponding to each storage location type;

[0048] According to the mean, the target storage location type is determined, and then the target storage location among the candidate storage locations is determined according to the target storage location type.

[0049] Optionally, the target storage location determination unit is further configured to:

[0050] Determine a storage location among the candidate storage locations that corresponds to the target storage location type;

[0051] The configuration data of the storage location corresponding to the target storage location type is obtained, and the target storage location is determined according to the attribute data and configuration data of the items to be put on the shelves.

[0052] Optionally, the storage location selection device further includes a model training unit, which is further configured to:

[0053] Acquire a sample set, wherein the sample set is configured as feature data of preset dimensions of historical items on the shelves, feature data of preset dimensions of historical items off the shelves, and corresponding sample storage location types;

[0054] Model training is performed based on the feature data of preset dimensions of historical items on the shelves, the feature data of preset dimensions of historical items off the shelves, and the corresponding sample storage location types to obtain a storage location classification selection model.

[0055] Optionally, the model training unit is further configured to:

[0056] Generate a corresponding decision tree based on the input feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves;

[0057] The generated decision trees are used to form a storage location classification selection model.

[0058] Optionally, the model training unit is further configured to:

[0059] Determine the information entropy corresponding to the feature data of the preset dimension of the historical listing items and the feature data of the preset dimension of the historical delisting items;

[0060] Target feature data is obtained by screening according to information entropy, and the target feature data is combined based on preset dimensions to obtain combined target feature data;

[0061] Generate a corresponding decision tree based on the combined target feature data.

[0062] In addition, the present application also provides a storage location determination electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by one or more processors, the one or more processors implement the storage location determination method as described above.

[0063] In addition, the present application also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the storage location determination method as described above is implemented.

[0064] One embodiment of the above invention has the following advantages or beneficial effects: the present application receives a storage request for items to be put on the shelves, obtains the attribute data of the items to be put on the shelves and the historical data of the candidate storage locations; performs cluster analysis on the historical data to determine the corresponding storage location type; extracts the corresponding dimension feature data in the attribute data of the items to be put on the shelves based on the preset dimension, and determines the corresponding dimension score according to the dimension feature data; determines and outputs the target storage location among the candidate storage locations according to the dimension score and the storage location type. The efficiency of processing items and the accuracy of selecting storage locations are improved.

[0065] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific implementation examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings are used to better understand the present application and do not constitute an improper limitation on the present application.

[0067] Figure 1 is a schematic diagram of the main process of a storage location determination method provided according to an embodiment of the present application;

[0068] Figure 2 is a schematic diagram of the main process of a storage location determination method provided according to an embodiment of the present application;

[0069] Figure 3 This is a schematic diagram of an application scenario of a storage location determination method provided according to an embodiment of the present application;

[0070] Figure 4 is a schematic diagram of the main units of the storage position determination device according to an embodiment of the present application;

[0071] Figure 5 is an exemplary system architecture diagram to which the embodiments of the present application can be applied;

[0072] Figure 6 It is a structural diagram of a computer system of a terminal device or server suitable for implementing an embodiment of the present application. DETAILED DESCRIPTION

[0073] The following is an explanation of the exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description. It should be noted that in the technical solution of the present disclosure, the collection, collection, update, analysis, processing, use, transmission, storage and other aspects of user personal information involved are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Take necessary measures for user personal information to prevent illegal access to user personal information data and maintain user personal information security, network security and national security.

[0074] Figure 1 is a schematic diagram of the main process of the storage location determination method provided according to an embodiment of the present application, such as Figure 1 As shown, the storage location determination method includes:

[0075] Step S101, receiving a storage request for items to be put on the shelves, and obtaining corresponding attribute data of the items to be put on the shelves and historical data of candidate storage locations.

[0076] In this embodiment, the execution subject of the storage location determination method (for example, it can be a server) can receive a storage request for items to be put on the shelves through a wired connection or a wireless connection. After receiving the storage request for items to be put on the shelves, the execution subject can obtain the attribute data of the items to be put on the shelves placed in the storage location. Specifically, the attribute data of the items to be put on the shelves may include data such as the weight, volume, storage conditions, quantity, etc. of the items to be put on the shelves. The embodiment of the present application does not specifically limit the attribute data of the items to be put on the shelves. The execution subject can also obtain the historical data of the candidate storage location. Specifically, the historical data of the candidate storage location may include the historical data of the items on the shelves of the candidate storage location, the historical data of the items off the shelves of the candidate storage location, and the type data of the items stored in the candidate storage location, etc. The embodiment of the present application does not specifically limit the historical data of the candidate storage location.

[0077] Step S102: performing cluster analysis on historical data to determine corresponding storage location types.

[0078] According to the type data of the items stored in the candidate storage locations in the historical data, the storage locations are classified to obtain the storage location type corresponding to each storage location. For example, the storage location type may include ordinary storage location, warm layer storage location, manual picking storage location, automatic picking storage location or zero picking storage location.

[0079] Specifically, cluster analysis is performed on historical data to determine corresponding storage location types, including: extracting corresponding storage location feature data from historical data based on preset dimensions; and inputting the storage location feature data into a cluster analysis model to output corresponding storage location types.

[0080] Among them, the cluster analysis model is determined by the following steps: obtaining sample feature data of preset dimensions and the storage location type corresponding to the sample feature data; using the sample feature data as the input of the initial cluster analysis model, and using the storage location type corresponding to the sample feature data as the expected output of the initial cluster analysis model, so as to train the initial cluster analysis model and generate the final cluster analysis model. Specifically, the sample feature data may include feature data of multiple dimensions of multiple historical storage location information, attribute data of historical items on and off the shelves, etc., and the embodiment of the present application does not specifically limit the sample feature data. The historical storage location information may include the type of storage location for historical storage items, the size of the storage location, etc., wherein the type of storage location may be, for example, an ordinary storage location, a temperature layer storage location, a manual picking storage location, an automated picking storage location, etc., and the embodiment of the present application does not specifically limit the storage location type. The initial cluster analysis model may be an initial neural network model, such as a convolutional neural network model (CNN), and the embodiment of the present application does not specifically limit the initial cluster analysis model.

[0081] The preset dimension may include, for example, one or more of a storage location size dimension, a storage location type dimension, a picking cost dimension, an item turnover rate dimension, and an item type dimension. The execution subject may extract corresponding storage location feature data from the historical data based on the preset dimension, such as one or more of storage location size feature data, storage location type feature data, picking cost feature data, item turnover rate feature data, and item type feature data. The embodiment of the present application does not specifically limit the storage location feature data.

[0082] Specifically, the storage location feature data obtained based on the historical data can be input into the cluster analysis model, so as to classify the historical data based on the pre-trained cluster analysis model to obtain the storage location type corresponding to each candidate storage location.

[0083] Step S103, extracting the corresponding dimension feature data from the attribute data of the items to be put on the shelves based on the preset dimension, and determining the corresponding dimension score according to the dimension feature data.

[0084] The preset dimensions in step S103 may include the quality dimension, size dimension, storage condition dimension, quantity dimension, etc. of the items to be put on the shelves. The embodiments of the present application do not specifically limit the preset dimensions. For example, the execution subject may extract one or more of the quality feature data, size feature data, storage condition feature data, and quantity feature data from the attribute data of the items to be put on the shelves based on the preset dimensions. Then, the execution subject may determine the corresponding dimension score based on the extracted dimension feature data.

[0085] Specifically, determining the corresponding dimension score includes: inputting the dimension feature data into the storage location classification selection model, and obtaining the dimension score of the corresponding preset dimension through the decision tree in the storage location classification selection model.

[0086] Specifically, before inputting the dimensional feature data into the storage location classification selection model, the storage location determination method also includes a model training step, specifically: obtaining a sample set, wherein the sample set includes feature data of preset dimensions of historically on-shelf items, feature data of preset dimensions of historically off-shelf items, and corresponding sample storage location types; performing model training based on the feature data of preset dimensions of historically on-shelf items, feature data of preset dimensions of historically off-shelf items, and corresponding sample storage location types to obtain a storage location classification selection model.

[0087] Specifically, the feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves in the sample set can be used as the input of the model to be trained, and the sample storage location types corresponding to the feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves can be used as the expected output of the model to be trained, and the model to be trained is trained to obtain a storage location classification selection model. The embodiment of the present application does not specifically limit the number of iterations and training termination conditions in the model training process.

[0088] Specifically, the storage location classification selection model is obtained, including: generating corresponding decision trees according to the input feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves; and forming the storage location classification selection model with the generated decision trees. Thus, the classification selection model composed of multiple decision trees can improve the accuracy of classification decisions.

[0089] Specifically, generating a corresponding decision tree includes: determining the information entropy corresponding to the feature data of preset dimensions of historically put-on-shelf items and the feature data of preset dimensions of historically taken-off-shelf items; obtaining target feature data by screening according to the information entropy, combining the target feature data based on preset dimensions to obtain combined target feature data; and generating a corresponding decision tree according to the combined target feature data. The embodiment of the present application generates a corresponding decision tree based on information entropy and the target feature data screened by the information entropy, and the decision tree generated can improve the accuracy of classification decisions.

[0090] In the embodiment of the present application, information entropy can be the probability of occurrence of feature data of preset dimensions of historical items on shelves and feature data of preset dimensions of historical items off shelves in the storage location selection scenario of the embodiment of the present application. The feature data corresponding to the information entropy being greater than the preset threshold can be determined as target feature data, thereby realizing the screening of target feature data. When there are multiple target feature data, the target feature data corresponding to each preset dimension is determined. When there are multiple target feature data corresponding to each preset dimension, the priority of the target feature data corresponding to each preset dimension is determined, and the corresponding target feature data is combined according to the priority to obtain the combined target feature data. For example, the target feature data corresponding to preset dimension 1 includes: target feature data 1, target feature data 2, and target feature data 3, and the corresponding priority is target feature data 2> target feature data 1> target feature data 3. Then the combined feature data can be target feature data 2-target feature data 1-target feature data 3. After obtaining the combined target feature data corresponding to each preset dimension, the execution subject can generate a corresponding decision tree based on each obtained combined target feature data. That is, each preset dimension corresponds to a decision tree.

[0091] Step S104, determining a target storage location among candidate storage locations according to the dimension score and storage location type and outputting the target storage location.

[0092] The dimension scores of preset dimensions are summarized according to the storage location type to determine the dimension scores corresponding to each storage location type, that is, the dimension scores of the preset dimensions corresponding to each storage location type are summarized together. The dimension score can be an average score, or it can be the sum of the scores corresponding to the preset dimensions, or it can be the mean difference or standard deviation of the scores corresponding to the preset dimensions. The embodiment of the present application does not specifically limit the dimension score. The preset dimensions, for example, can be dimensions such as weight, volume, category and picking rate of items to be put on the shelves. The embodiment of the present application does not specifically limit the preset dimensions. Specifically, one or more storage locations corresponding to the storage location type with the highest dimension score can be determined as the target storage location and output to the terminal corresponding to the user.

[0093] This embodiment receives a storage request for items to be put on the shelves, obtains the corresponding attribute data of the items to be put on the shelves and the historical data of the candidate storage locations; performs cluster analysis on the historical data to determine the corresponding storage location type; extracts the corresponding dimension feature data from the attribute data of the items to be put on the shelves based on the preset dimension, and determines the corresponding dimension score according to the dimension feature data; determines and outputs the target storage location among the candidate storage locations according to the dimension score and the storage location type, thereby improving the efficiency of processing items and the accuracy of selecting storage locations.

[0094] Figure 2FIG. 1 is a schematic diagram of the main flow of a storage location determination method provided according to an embodiment of the present application. Figure 2 As shown, the storage location determination method includes:

[0095] Step S201, receiving a storage request for items to be put on the shelves, and obtaining attribute data of the items to be put on the shelves and historical data of candidate storage locations.

[0096] The storage request for items to be put on the shelves may be a request to select the most suitable storage location for the items to be put on the shelves. The most suitable storage location may specifically be the storage location with the smallest remaining space after the items to be put on the shelves are placed into the storage location. The attribute data of the items to be put on the shelves may include, for example, data such as the material and stability of the items to be put on the shelves. The embodiment of the present application does not specifically limit the attribute data of the items to be put on the shelves. The historical data of the candidate storage locations may include data such as the material of the items that have been stored in the candidate storage locations in the past, the stability of the stored items, the types of items that have been stored in the candidate storage locations in the past, and the turnover rate / picking rate of the stored items.

[0097] Step S202: Perform cluster analysis on historical data to determine corresponding storage location types.

[0098] Based on the types of stored items, cluster analysis is performed on the candidate storage locations corresponding to the historical data to obtain the corresponding storage location types.

[0099] Step S203, extracting the corresponding dimension feature data from the attribute data of the items to be put on the shelves based on the preset dimension, and determining the corresponding dimension score according to the dimension feature data.

[0100] Specifically, the dimension feature data may be input into a storage location classification selection model to calculate the dimension score corresponding to each dimension feature data.

[0101] Step S204 , summarizing the dimension scores of the preset dimensions according to the storage location types to determine the dimension score corresponding to each storage location type.

[0102] A storage location type may correspond to multiple dimensional feature data, that is, a storage location type may correspond to multiple dimensional scores. Determine and summarize one or more dimensional scores corresponding to each storage location type.

[0103] Step S205, calculating the mean of the dimension scores corresponding to each storage location type.

[0104] Specifically, the weight value corresponding to the dimensional feature data corresponding to each storage type can be obtained, and then the sum of the dimensional scores corresponding to each storage type can be calculated based on the weight value. For example, the weight value corresponding to dimensional feature data 1 is a1, the weight value corresponding to dimensional feature data 2 is a2, and the weight value corresponding to dimensional feature data 3 is a3, and the sum is: a1*dimensional feature data 1+a2*dimensional feature data 2+a3*dimensional feature data 3=k. And the average of the sum is calculated to obtain the mean value k / 3 of the dimensional score corresponding to each storage type.

[0105] Of course, as another implementation of the embodiment of the present application, the dimension scores corresponding to each storage location type may be directly summed and averaged to obtain a mean value. For example, the sum is: dimension feature data 1 + dimension feature data 2 + dimension feature data 3 = h. And the sum is averaged to obtain the mean value h / 3 of the dimension scores corresponding to each storage location type.

[0106] Step S206, determining the target storage location type according to the mean value, and then determining the target storage location among the candidate storage locations according to the target storage location type.

[0107] The storage location type corresponding to when the mean value exceeds the preset threshold value may be determined as the target storage location type. The preset threshold value is not specifically limited in the present embodiment of the application.

[0108] Specifically, determining a target storage location among candidate storage locations according to the target storage location type includes: determining a storage location among the candidate storage locations corresponding to the target storage location type; acquiring configuration data of the storage location corresponding to the target storage location type, and determining the target storage location according to attribute data and configuration data of items to be put on the shelves.

[0109] When the target storage location type corresponds to multiple candidate storage locations, the execution entity can screen out the final target storage location based on the preset configuration data for each candidate storage location and the data of items to be put on the shelves.

[0110] Configuration data of each candidate storage location, such as how many items can be placed, whether items can be mixed, etc. The embodiment of the present application does not specifically limit the configuration data of each candidate storage location. Specifically, based on the weight, size, storage conditions, whether items can be mixed, etc. of the data of the items to be put on the shelves, as well as the weight restrictions, size restrictions, storage conditions restrictions, and whether items can be mixed in the configuration data of each candidate storage location, the candidate storage location that meets the storage conditions of the items to be put on the shelves can be selected as the target storage location.

[0111] Furthermore, if there is no candidate storage location that 100% meets the requirements for shelving items among the candidate storage locations, the candidate storage location with a matching degree greater than a preset threshold (e.g., 90%) can be determined as the target storage location. Specifically, if the weight of the data of the item to be shelved is within the weight limit of the candidate storage location, it can be determined that there is a matching data between the data of the item to be shelved and the quality inspection of the candidate storage location, and the size, storage conditions, and whether the data of the item to be shelved can be mixed are compared with the configuration data of the candidate storage location to determine the number of matching data N and the number of matching data n, and the matching degree is n / N.

[0112] Therefore, the embodiments of the present application can accurately and quickly recommend storage locations for items to be put on the shelves, thereby improving the efficiency and accuracy of storage location selection.

[0113] Figure 3 The diagram is a schematic diagram of an application scenario of a storage location determination method provided in an embodiment of the present application. The storage location determination method in the embodiment of the present application can be applied to a scenario of putting items on shelves.

[0114] Step S301, obtaining historical multiple storage location multiple dimensional feature data.

[0115] Obtain feature data of multiple dimensions of multiple historical storage location information and train a cluster analysis model.

[0116] Step S302: training a cluster analysis model to obtain storage location categories corresponding to the storage locations.

[0117] The data of storage locations and storage areas can be classified according to one or more dimensions such as different storage location sizes, different storage location types, different picking costs, different item turnover rates / picking rates, and different item types to obtain storage location categories. Among them, different storage location sizes identify storage locations with different volume sizes. Different storage location types include ordinary storage locations, temperature-layer storage locations, manual picking storage locations, automated picking storage locations, or storage storage locations, zero-pick storage locations, etc. Different picking costs identify the distance of the storage location from the entrance and exit and the level of the storage location.

[0118] C=∑(D in +D out +H in +H out )

[0119] Among them, C is the storage information, D in is the distance from the storage location to the entrance and exit when the items are put on the shelf, D out H is the distance from the storage location to the entrance and exit when the items are removed from the shelf. in is the cost of different levels when the storage items are put on the shelf, H out It is the different levels of cost when storage items are taken off the shelves.

[0120] The level cost of the same shelf is different for different levels of storage location picking costs. For example, the first level is 0, and the level cost increases by 1 for each additional level. Different item turnover rates / picking rates identify the frequency of picking at a storage location within a certain period of time, which is usually obtained based on the mathematical statistics of historical data. Different item types identify the types of items that can be put on the shelves at a storage location. The clustering algorithm model is trained based on the multiple dimensional feature data corresponding to the multiple historical storage location information and historical item data obtained above.

[0121] Specifically, the information of multiple historical storage locations and multiple dimensional feature data of storage location storage items are selected, and cluster analysis can be performed using the k-means algorithm, DBSCAN algorithm, etc. to divide the storage locations into multiple categories. Storage locations in the same group have similar attributes and / or characteristics, while storage locations in different groups have highly different attributes and / or characteristics.

[0122] Acquire multiple dimensional feature data of the attribute data of the items to be put on the shelves. Input the acquired multiple dimensional feature data of the attribute data of the items to be put on the shelves into a pre-trained storage location classification selection model to determine the storage location to be put on the shelves of the attribute data of the items to be put on the shelves.

[0123] Among them, the trained storage location classification selection model can be trained in the following way: obtain a large amount of historical warehouse item data and the storage area and storage location information corresponding to each item, wherein the warehouse item data, storage area and storage location information include feature data of multiple dimensions, and perform random forest model training based on the feature data of multiple dimensions of a large amount of historical warehouse item data, storage area and storage location data to obtain the item storage location classification selection model.

[0124] Obtain multiple dimensional feature data of multiple historical warehoused item data and storage location data for model training to obtain a model for item storage location classification and selection. The multiple dimensional feature data of warehoused items mainly include dimensions such as storage space, weight, and item type. The dimensional types may involve dozens or even hundreds of types, which are not specifically limited here.

[0125] Based on the storage location classification obtained by the clustering algorithm, the historical information of items on the shelves and historical information of items off the shelves are added with the storage location category dimension to obtain the storage location category corresponding to each historical item as the input of the random forest model.

[0126] Step S303, training a random forest model based on multiple dimensional feature data of historical incoming items, out-of-shelf items, and storage location categories.

[0127] The random forest model is trained using multi-dimensional feature data of multiple historical inventory item data, multi-dimensional feature data of delisted item data, and storage location category data corresponding to each historical item data. The various parameters in the random forest model, such as the number of trees in the forest, the depth of the tree, and the number of leaf nodes, are determined. Information entropy or Gini index is used as the basis for judgment, thereby obtaining a storage location classification selection model.

[0128] Specifically, one or more of the multiple dimensional feature data such as length, width, height, category, temperature layer, volume, turnover rate, etc. in the historical data of multiple items on the shelves can be obtained as input, and the storage location classification information corresponding to the historical items on the shelves is used as the expected output to train the decision tree model. If there are multiple dimensional features of the input training sample, the best feature is selected according to the information gain / information gain ratio / Gini index to generate a decision tree. After the above training, the decision tree model corresponding to each / multiple dimensional combination features is obtained, and the generated multiple decision trees are combined into a random forest model, and the final classification result is determined by voting on multiple decision trees. In the embodiment of the present application, the random forest model is the final storage location classification selection model.

[0129] Step S304: input the information of the newly stored items into the random forest model to predict the storage location category.

[0130] The feature data of multiple dimensions of the items to be put on the shelves and the feature data of multiple dimensions of the storage locations are input into a pre-trained storage location classification selection model to determine the storage location category of the items to be put on the shelves.

[0131] Specifically, the characteristic data of each dimension of the above-mentioned items to be put on the shelves and storage locations can be judged against the decision tree characteristic data range of each dimension in the storage location classification selection model, so as to match the storage location category selection result corresponding to the obtained dimensional characteristic data range as the storage location category selection result corresponding to each dimension of the items to be put on the shelves, that is, the selection probability of the shelf item data for the above-mentioned storage location categories under this dimension; then, according to the storage location category selection results corresponding to each dimension of the items to be put on the shelves, the selection probability of the overall item for each storage location category under multiple dimensions is determined, and then the one with the largest mean selection probability is determined from each storage location category as the storage location category where the items to be put on the shelves should be placed.

[0132] Step S305: Select a recommended storage location from a set of storage locations that meet the storage location category according to the shelf restriction rules.

[0133] The storage location categories corresponding to the items to be put on the shelves are predicted by the random forest model, and a set of storage locations that can meet the requirements of putting the items on the shelves and belong to this storage location category are screened out.

[0134] Step S306: putting items on the shelves.

[0135] Step S307, completing the listing.

[0136] The above storage location set is selected according to certain configuration rules. For example, after the items are put on the shelves, the available volume of the storage location is the smallest. Finally, the recommended storage location for the items to be put on the shelves is obtained.

[0137] For example, the feature data of multiple dimensions of storage locations and the feature data of multiple dimensions of items on and off shelves listed in the examples of this application are only used for illustration, and the actual application is more complicated, as shown below:

[0138]

[0139] For example, the decision trees corresponding to each dimension are trained based on the feature data of each dimension of weight, volume, picking rate and category in multiple historical data of items on the shelves and the storage location category data corresponding to each historical data of items on the shelves, such as decision trees 1-4. Then, the decision trees corresponding to each dimension are combined to obtain a random forest model, that is, a storage location classification selection model.

[0140] Decision Tree 1:

[0141] Weight(kg) A B C D E ≤1 30% 35% 25% 10% 0% 1-2 10% 60% 25% 5% 0% 2-3 40% 40% 15% 5% 0% ≥3 5% 5% 70% 20% 0%

[0142] Decision Tree 2:

[0143] <![CDATA[Volume (cm 3 )]]> A B C D E ≤100 40% 40% 15% 5% 0% 100-500 5% 5% 70% 20% 0% 500-1000 20% 20% 50% 10% 0% ≥1000 35% 30% 25% 10% 0%

[0144] Decision Tree 3:

[0145]

[0146]

[0147] Decision Tree 4:

[0148] Category A B C D E Digital 5% 20% 60% 15% 0% food 15% 15% 50% 20% 0% Shoes and Clothing 50% 40% 5% 5% 0% Fresh 0% 0% 0% 0% 100%

[0149] The characteristic data of the weight, volume, category and picking rate of the items to be put on the shelves, for example, 1.5kg, 800cm 3, digital, digital, match the feature data of each dimension in the target order data with the feature data range of the corresponding dimension in the cluster analysis model, and obtain the decision results corresponding to each dimension of the attribute data of the target items to be put on the shelves, that is, the decision probability of various storage location types for the feature data range of each dimension of the items, and then calculate the decision probability of each storage location type classification corresponding to the attribute data of the items to be put on the shelves according to the random forest model, that is, the storage location classification selection model. For example, the mean of the decision probability of each storage location type under each dimension can be calculated. For example, as shown in the following table of the random forest model, the storage location type C with the largest average value of the decision probability is taken as the storage location type corresponding to the attribute data of the items to be put on the shelves.

[0150] Random Forest Model:

[0151]

[0152] The embodiment of the present application determines the probability of putting an item on the shelf in a certain storage location based on multiple dimensions of the item (such as weight, category, whether it is a room temperature item, freshness, volume, turnover rate, etc.), and selects a storage location category with the highest average probability through a decision tree for each dimension.

[0153] After obtaining the storage location classification, the storage location set that can be put on the shelf is obtained in the above classification according to the storage location restriction conditions (such as how many items can be placed, whether items can be mixed, etc.), and the optimal storage location is selected from the above set. The embodiment of the present application can improve the efficiency of processing items and the accuracy of storage location selection.

[0154] Figure 4 Schematic diagram of the main units of the storage position determination device according to an embodiment of the present application. Figure 4 As shown, the storage location determination device 400 includes a receiving unit 401, a cluster analysis unit 402, a dimension score determination unit 403 and a target storage location determination unit 404.

[0155] The receiving unit 401 is configured to receive a storage request for items to be put on the shelves, and obtain the corresponding attribute data of the items to be put on the shelves and the historical data of the candidate storage locations.

[0156] The cluster analysis unit 402 is configured to perform cluster analysis on the historical data to determine the corresponding storage location type.

[0157] The dimension score determination unit 403 is configured to extract corresponding dimension feature data from the attribute data of the items to be put on the shelves based on the preset dimension, and determine the corresponding dimension score according to the dimension feature data.

[0158] The target storage location determination unit 404 is configured to determine and output a target storage location among the candidate storage locations according to the dimension score and the storage location type.

[0159] In some embodiments, the clustering analysis unit 402 is further configured to: extract corresponding storage location feature data in historical data based on preset dimensions; input the storage location feature data into the clustering analysis model to output the corresponding storage location type, wherein the clustering analysis model is determined by the following steps: obtaining sample feature data of preset dimensions and the storage location type corresponding to the sample feature data; using the sample feature data as the input of the initial clustering analysis model, and using the storage location type corresponding to the sample feature data as the expected output of the initial clustering analysis model, so as to train the initial clustering analysis model and generate a final clustering analysis model.

[0160] In some embodiments, the dimension score determination unit 403 is further configured to: input the dimension feature data into the storage location classification selection model, and obtain the dimension score of the corresponding preset dimension through the decision tree in the storage location classification selection model.

[0161] In some embodiments, the target storage location determination unit 404 is further configured to: summarize the dimension scores of preset dimensions according to the storage location type to determine the dimension scores corresponding to each storage location type; calculate the mean of the dimension scores corresponding to each storage location type; determine the target storage location type according to the mean, and then determine the target storage location among the candidate storage locations according to the target storage location type.

[0162] In some embodiments, the target storage location determination unit 404 is further configured to: determine a storage location among the candidate storage locations corresponding to the target storage location type; obtain configuration data of the storage location corresponding to the target storage location type, and determine the target storage location based on the attribute data and configuration data of the items to be put on the shelves.

[0163] In some embodiments, the storage location selection device further includes Figure 4 The model training unit not shown in the figure is further configured to: obtain a sample set, wherein the sample set is configured into feature data of preset dimensions of historical items on the shelves, feature data of preset dimensions of historical items off the shelves, and corresponding sample storage location types; perform model training based on the feature data of preset dimensions of historical items on the shelves, feature data of preset dimensions of historical items off the shelves, and corresponding sample storage location types to obtain a storage location classification selection model.

[0164] In some embodiments, the model training unit is further configured to: generate a corresponding decision tree based on the input feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves; and form a storage location classification selection model with the generated decision tree.

[0165] In some embodiments, the model training unit is further configured to: determine the information entropy corresponding to the feature data of preset dimensions of historical items on the shelves and the feature data of preset dimensions of historical items off the shelves; obtain target feature data based on the information entropy screening, and combine the target feature data based on the preset dimensions to obtain combined target feature data; generate a corresponding decision tree based on the combined target feature data.

[0166] The execution subject may also be provided with a setting module for setting basic configuration information, including the configuration of restrictions on the storage locations to be put on the shelves; a collection module: collecting the data of the warehouse management system for putting on the shelves, taking off the shelves, item data, and storage location data, and reporting them to the background database; a cluster analysis training module: using cluster learning algorithms to calculate the classification categories of storage locations; a storage location selection module for putting on the shelves: screening the storage location set under the storage location category of the items to be put on the shelves, and recommending the storage locations for putting on the shelves according to certain rules. A shelf allocation module: performing the putting on the shelves operation according to the recommended storage locations for putting on the shelves.

[0167] It should be noted that the storage location determination method and storage location determination device of the present application have a corresponding relationship in terms of specific implementation contents, so the repeated contents will not be described again.

[0168] Figure 5 An exemplary system architecture 500 is shown to which the storage location determination method or storage location determination device according to the embodiments of the present application can be applied.

[0169] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0170] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 501, 502, 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0171] The terminal devices 501, 502, 503 may be various electronic devices having a storage location selection processing screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0172] Server 505 may be a server that provides various services, such as a background management server that provides support for storage requests for items to be put on the shelves submitted by users using terminal devices 501, 502, and 503 (only as an example). The background management server may receive storage requests for items to be put on the shelves, obtain the corresponding attribute data of the items to be put on the shelves and the historical data of the candidate storage locations; perform cluster analysis on the historical data to determine the corresponding storage location type; extract the corresponding dimension feature data in the attribute data of the items to be put on the shelves based on the preset dimensions, and determine the corresponding dimension score according to the dimension feature data; determine the target storage location among the candidate storage locations according to the dimension score and the storage location type and output it. This improves the efficiency of processing items and the accuracy of storage location selection.

[0173] It should be noted that the storage location determination method provided in the embodiment of the present application is generally executed by the server 505, and accordingly, the storage location determination device is generally set in the server 505.

[0174] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0175] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present application. Figure 6 The terminal device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0176] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the computer system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0177] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0178] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present application are executed.

[0179] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. Computer-readable storage media may include, for example, but are not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0180] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0181] The units involved in the embodiments described in the present application may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes a receiving unit, a clustering analysis unit, a dimension score determination unit, and a target storage location determination unit. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0182] As another aspect, the present application also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device receives a storage request for items to be put on the shelves, obtains the corresponding attribute data of the items to be put on the shelves and the historical data of the candidate storage locations; performs cluster analysis on the historical data to determine the corresponding storage location type; extracts the corresponding dimensional feature data in the attribute data of the items to be put on the shelves based on the preset dimension, and determines the corresponding dimensional score according to the dimensional feature data; determines the target storage location among the candidate storage locations according to the dimensional score and the storage location type and outputs it.

[0183] According to the technical solution of the embodiment of the present application, the efficiency of processing items and the accuracy of storage location selection can be improved.

[0184] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A storage location determination method, It is characterized in that include: Receiving a storage request for items to be put on the shelves, and obtaining attribute data of the items to be put on the shelves and historical data of candidate storage locations; Performing cluster analysis on the historical data to determine corresponding storage location types; Extracting corresponding dimension feature data from the attribute data of the items to be put on the shelves based on the preset dimensions, and determining corresponding dimension scores according to the dimension feature data; According to the dimension score and the storage location type, a target storage location among the candidate storage locations is determined and output.

2. The method according to claim 1, It is characterized in that The cluster analysis of the historical data to determine the corresponding storage location type includes: Extracting corresponding storage location characteristic data from the historical data based on a preset dimension; The storage location characteristic data is input into a cluster analysis model to output a corresponding storage location type, wherein the cluster analysis model is determined by the following steps: Acquire sample feature data of a preset dimension and a storage location type corresponding to the sample feature data; The sample characteristic data is used as the input of the initial cluster analysis model, and the storage location type corresponding to the sample characteristic data is used as the expected output of the initial cluster analysis model, so as to train the initial cluster analysis model and generate a final cluster analysis model.

3. The method according to claim 1, It is characterized in that The determining the corresponding dimension score includes: The dimension feature data is input into the storage location classification selection model, and the dimension score of the corresponding preset dimension is obtained through the decision tree in the storage location classification selection model.

4. The method according to claim 1, It is characterized in that The step of determining a target storage location among the candidate storage locations according to the dimension score and the storage location type includes: Summarizing the dimension scores of the preset dimensions according to the storage location types to determine the dimension score corresponding to each storage location type; Calculate the mean of the dimension scores corresponding to each of the storage location types; According to the mean, a target storage location type is determined, and then according to the target storage location type, a target storage location among the candidate storage locations is determined.

5. The method according to claim 4, It is characterized in that The step of determining a target storage location among the candidate storage locations according to the target storage location type includes: Determine a storage location among the candidate storage locations that corresponds to the target storage location type; The configuration data of the storage location corresponding to the target storage location type is obtained, and the target storage location is determined according to the attribute data of the item to be put on the shelf and the configuration data.

6. The method according to claim 3, It is characterized in that Before inputting the dimensional feature data into the storage location classification selection model, the method further includes: Acquire a sample set, wherein the sample set includes feature data of preset dimensions of historical items on the shelves, feature data of preset dimensions of historical items off the shelves, and corresponding sample storage location types; Model training is performed based on the feature data of preset dimensions of historical items on the shelves, the feature data of preset dimensions of historical items off the shelves, and the corresponding sample storage location types to obtain a storage location classification selection model.

7. The method according to claim 6, It is characterized in that The method of obtaining the storage location classification selection model comprises: Generate a corresponding decision tree according to the input feature data of preset dimensions of the historical items on the shelves and the feature data of preset dimensions of the historical items off the shelves; The generated decision tree is used to form a storage location classification selection model.

8. The method according to claim 7, It is characterized in that The generating a corresponding decision tree comprises: Determine the information entropy corresponding to the feature data of the preset dimension of the historical items on the shelves and the feature data of the preset dimension of the historical items off the shelves; Target feature data is obtained by screening according to the information entropy, and the target feature data is combined based on a preset dimension to obtain combined target feature data; A corresponding decision tree is generated according to the combined target feature data.

9. A storage position determination device, It is characterized in that include: A receiving unit, configured to receive a storage request for items to be put on the shelves, and obtain attribute data of the items to be put on the shelves and historical data of candidate storage locations; a cluster analysis unit configured to perform cluster analysis on the historical data to determine a corresponding storage location type; A dimension score determination unit is configured to extract corresponding dimension feature data from the attribute data of the to-be-shelved item based on a preset dimension, and determine a corresponding dimension score according to the dimension feature data; The target storage location determination unit is configured to determine and output a target storage location among the candidate storage locations based on the dimension score and the storage location type.

10. The device according to claim 9, It is characterized in that The cluster analysis unit is further configured to: Extracting corresponding storage location characteristic data from the historical data based on a preset dimension; The storage location characteristic data is input into a cluster analysis model to output a corresponding storage location type, wherein the cluster analysis model is determined by the following steps: Acquire sample feature data of a preset dimension and a storage location type corresponding to the sample feature data; The sample characteristic data is used as the input of the initial cluster analysis model, and the storage location type corresponding to the sample characteristic data is used as the expected output of the initial cluster analysis model, so as to train the initial cluster analysis model and generate a final cluster analysis model.

11. An electronic device for determining a storage location, It is characterized in that include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

12. A computer readable medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.