A method and device for determining storage classification of items

By constructing a regression analysis model and clustering algorithm, the number and boundaries of item classification are automatically determined based on the historical outbound volume of items, the subjectivity and complexity of item storage classification in the existing technology are solved, and multi-category division and efficient calculation are realized .

CN114841647BActive Publication Date: 2024-11-19BEIJING JINGDONG SHANGKE INFORMATION TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210514524.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-12
Publication Date
2024-11-19
Estimated Expiration
2042-05-12

AI Technical Summary

Technical Problem

In the prior art, the number of items storage classifications and classification boundaries are subjectively determined, and the algorithm model-based method is highly complex when solving parameters, making it difficult to divide more categories.

Method used

By calculating the cumulative proportion of the inventory and the cumulative proportion of the item items based on the historical outbound volume of the items, a regression analysis model is constructed to determine the classification quantity and boundaries, a clustering algorithm and k-means algorithm are used to automatically determine the classification quantity, and Multi-category division is achieved through equidistant division.

Benefits of technology

It realizes item classification driven by actual data, simplifies the calculation of classification boundaries, overcomes the problems of subjective determination and high complexity, supports multi-category division, and improves classification accuracy and computing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114841647B_ABST
    Figure CN114841647B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and device for determining the storage classification of items, and relates to the field of warehousing technology. A specific implementation of the method includes: determining the number of classifications of each item according to the historical outbound quantity of each item; calculating the cumulative proportion of outbound quantity and the cumulative proportion of the number of items of each item according to the historical outbound quantity of each item; solving the parameters in the regression analysis model based on the cumulative proportion of outbound quantity and the cumulative proportion of the number of items of each item, thereby obtaining a classification model; calculating the boundaries of each classification based on the number of classifications and the classification model, thereby determining the classification of each item. This implementation can solve the technical problem of subjectively determining the number of classifications, classification boundaries, or solving the model parameters with high complexity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of warehousing technology, and in particular to a method and device for determining storage classification of items. Background Art

[0002] For the storage of items (hereinafter referred to as SKU), they are usually divided into several categories according to their outbound volume, and different categories of items are stored in different areas. This method is often called ABC classification in the industry. The specific steps include: according to the 80 / 20 rule, the items are divided into two categories, one is the items with the first 20% outbound volume, and the rest are the other category; or the items are divided into 3 categories with the help of an algorithm model, and then stored in categories.

[0003] The main disadvantages of existing methods are: (1) Determining the number of categories and classification boundaries of items is subjective rather than driven by actual data; (2) Although the division of classification boundaries is objective, the methods based on algorithmic models can only be divided into three categories and cannot be divided into more categories; (3) The methods based on algorithmic models need to solve nonlinear equations when calculating parameters, which are highly complex and difficult to solve. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and apparatus for determining storage classification of items to solve the technical problem of subjectively determining the number of classifications, classification boundaries, or solving model parameters with high complexity.

[0005] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for determining a storage classification of an item is provided, comprising:

[0006] Determine the classification quantity of each item according to the historical outbound quantity of each item;

[0007] Calculate the cumulative proportion of the shipment quantity and the cumulative proportion of the number of items of each item according to the historical shipment quantity of each item;

[0008] Solving the parameters in the regression analysis model based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, thereby obtaining a classification model;

[0009] Based on the number of categories and the classification model, the boundaries of each category are calculated to determine the category to which each item belongs.

[0010] Optionally, determining the number of categories of each item according to the historical outbound quantity of each item includes:

[0011] A clustering algorithm is used to cluster the items according to their historical outbound quantities, thereby determining the number of categories of the items.

[0012] Optionally, clustering the items according to the historical outbound quantity of each item by using a clustering algorithm, thereby determining the number of categories of each item, including:

[0013] For each clustering, the number of clusters is pre-specified, and the k-means algorithm is used to cluster the items according to the historical outbound volume of each item;

[0014] The inter-group spacing value or the intra-group spacing value of each clustering is compared, and the number of clusters corresponding to the cluster with the smallest inter-group spacing value or the smallest intra-group spacing value is selected as the classification number of each object.

[0015] Optionally, calculating the cumulative proportion of the shipment quantity of each item and the cumulative proportion of the number of items according to the historical shipment quantity of each item includes:

[0016] Sorting the items according to their historical shipment quantities, thereby obtaining a shipment quantity ranking of the items;

[0017] Calculate the cumulative proportion of the outbound quantity of each item according to the historical outbound quantity of each item;

[0018] The cumulative proportion of the number of items of each item is calculated according to the order of the outbound quantity of each item.

[0019] Optionally, solving the parameters in the regression analysis model based on the cumulative proportion of the shipment volume of each item and the cumulative proportion of the number of items, so as to obtain a classification model, includes:

[0020] Constructing a regression analysis model; wherein the regression analysis model describes the regression relationship between the cumulative proportion of the outbound volume and the cumulative proportion of the number of items;

[0021] Constructing an objective function according to the regression analysis model, and solving the objective function based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to minimize the difference between the fitted cumulative proportion of the outbound quantity and the actual cumulative proportion of the outbound quantity;

[0022] The parameters in the regression analysis model are obtained according to the solution result of the objective function, thereby obtaining a classification model.

[0023] Optionally, based on the number of categories and the classification model, calculating the boundaries of each category, thereby determining the category to which each item belongs, includes:

[0024] If the number of categories is 2, the cumulative proportion of the number of items at the inflection point of the classification model is calculated, so as to determine the category of each item according to the cumulative proportion of the number of items at the inflection point;

[0025] If the number of categories is 3, the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point are solved based on the classification model, so as to determine the category of each item according to the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point;

[0026] If the number of categories is greater than 3, the items are divided into three categories, and at least one of the three categories is divided equally, so as to determine the category to which the items belong.

[0027] Optionally, dividing at least one of the three categories into equal intervals to determine the category of each item includes:

[0028] Let m be the integer part of k / 3, and n be the remainder of k / 3; where k is the number of categories;

[0029] If n is equal to 1, the first category is divided into m+1 categories with equal intervals, and the second and third categories are divided into m categories with equal intervals; if n is equal to 2, the first and second categories are divided into m+1 categories with equal intervals, and the third category is divided into m categories with equal intervals;

[0030] The boundary points of each category are determined according to the equidistant division results, and the category of each item is determined according to the cumulative proportion of the number of items at the boundary points of each category.

[0031] In addition, according to another aspect of an embodiment of the present invention, there is provided a device for determining a storage classification of an item, comprising:

[0032] A determination module, used to determine the classification quantity of each item according to the historical outbound quantity of each item;

[0033] A calculation module, used to calculate the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items according to the historical outbound quantity of each item;

[0034] A solution module, used to solve the parameters in the regression analysis model based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to obtain a classification model;

[0035] The classification module is used to calculate the boundaries of each classification based on the classification quantity and the classification model, so as to determine the classification of each item.

[0036] Optionally, the determining module is further used for:

[0037] A clustering algorithm is used to cluster the items according to their historical outbound quantities, thereby determining the number of categories of the items.

[0038] Optionally, the determining module is further used for:

[0039] For each clustering, the number of clusters is pre-specified, and the k-means algorithm is used to cluster the items according to the historical outbound volume of each item;

[0040] The inter-group spacing value or the intra-group spacing value of each clustering is compared, and the number of clusters corresponding to the cluster with the smallest inter-group spacing value or the smallest intra-group spacing value is selected as the classification number of each object.

[0041] Optionally, the calculation module is further used for:

[0042] Sorting the items according to their historical shipment quantities, thereby obtaining a shipment quantity ranking of the items;

[0043] Calculate the cumulative proportion of the outbound quantity of each item according to the historical outbound quantity of each item;

[0044] The cumulative proportion of the number of items of each item is calculated according to the order of the outbound quantity of each item.

[0045] Optionally, the solution module is further used for:

[0046] Constructing a regression analysis model; wherein the regression analysis model describes the regression relationship between the cumulative proportion of the outbound volume and the cumulative proportion of the number of items;

[0047] Constructing an objective function according to the regression analysis model, and solving the objective function based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to minimize the difference between the fitted cumulative proportion of the outbound quantity and the actual cumulative proportion of the outbound quantity;

[0048] The parameters in the regression analysis model are obtained according to the solution result of the objective function, thereby obtaining a classification model.

[0049] Optionally, the classification module is further used for:

[0050] If the number of categories is 2, the cumulative proportion of the number of items at the inflection point of the classification model is calculated, so as to determine the category of each item according to the cumulative proportion of the number of items at the inflection point;

[0051] If the number of categories is 3, the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point are solved based on the classification model, so as to determine the category of each item according to the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point;

[0052] If the number of categories is greater than 3, the items are divided into three categories, and at least one of the three categories is divided equally, so as to determine the category to which the items belong.

[0053] Optionally, the classification module is further used for:

[0054] Let m be the integer part of k / 3, and n be the remainder of k / 3; where k is the number of categories;

[0055] If n is equal to 1, the first category is divided into m+1 categories with equal intervals, and the second and third categories are divided into m categories with equal intervals; if n is equal to 2, the first and second categories are divided into m+1 categories with equal intervals, and the third category is divided into m categories with equal intervals;

[0056] The boundary points of each category are determined according to the equidistant division results, and the category of each item is determined according to the cumulative proportion of the number of items at the boundary points of each category.

[0057] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, including:

[0058] one or more processors;

[0059] a storage device for storing one or more programs,

[0060] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of the above embodiments.

[0061] According to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0062] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0063] An embodiment of the above invention has the following advantages or beneficial effects: because the number of categories is determined according to the historical outbound quantity of each item, the parameters in the regression analysis model are solved based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, thereby obtaining the classification model, and finally the boundaries of each classification are calculated based on the number of categories and the classification model. Therefore, the technical problem of subjectively determining the number of categories, classification boundaries or solving the model parameters with high complexity in the prior art is overcome. The embodiment of the present invention can be driven by actual data instead of subjectively determining the number of categories and classification boundaries. The embodiment of the present invention first determines how many categories the items should be divided into based on the historical outbound quantity, and then quickly constructs the classification model through the regression analysis model and the historical outbound quantity, and quickly calculates the classification boundaries through the classification model. In the embodiment of the present invention, the number of categories can be unlimited, and it is simple and easy to solve, which is convenient for implementation.

[0064] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0066] Figure 1 is a schematic diagram of the main process of the method for determining the storage classification of items according to an embodiment of the present invention;

[0067] Figure 2 is a schematic diagram of the main process of a method for determining storage classification of items according to a reference embodiment of the present invention;

[0068] Figure 3 It is a schematic diagram of the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items according to an embodiment of the present invention;

[0069] Figure 4 is a schematic diagram of a classification model according to an embodiment of the present invention;

[0070] Figure 5 is a schematic diagram of main modules of an apparatus for determining storage classification of items according to an embodiment of the present invention;

[0071] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0072] Figure 7 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0073] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0074] Figure 1 Schematic diagram of the main process of the method for determining the storage classification of items according to an embodiment of the present invention. As an embodiment of the present invention, Figure 1 As shown, the method for determining the storage classification of items may include:

[0075] Step 101, determining the classification quantity of each item according to the historical delivery quantity of each item.

[0076] The embodiment of the present invention is driven by actual data and calculates the number of categories based on the historical outbound volume of each item (the outbound volume in a recent period of time), rather than subjectively determining the number of categories. Optionally, step 101 may include: clustering the items according to the historical outbound volume of each item using a clustering algorithm, thereby determining the number of categories of each item. In order to accurately calculate the number of categories, the embodiment of the present invention uses a clustering algorithm to cluster the items, and determines the number of categories of each item based on the clustering results, that is, the number of clusters is equal to the number of categories.

[0077] Optionally, a clustering algorithm is used to cluster the items according to the historical outbound quantity of each item, so as to determine the number of categories of each item, including: for each clustering, the number of clusters is pre-specified, and the k-means algorithm is used to cluster the items according to the historical outbound quantity of each item; the inter-group spacing value or the intra-group spacing value of each clustering is compared, and the number of clusters corresponding to the cluster with the smallest inter-group spacing value or the smallest intra-group spacing value is selected as the number of categories of each item. For example, the number of clusters must be specified in advance in k-means. Considering that the number of clusters is generally 2 to 6 in the warehouse storage scenario, the number of clusters is pre-selected as 2, 3, 4, 5, and 6, and these cluster numbers are clustered respectively, and then the inter-group spacing value or the intra-group spacing value of each clustering is compared, and the smallest inter-group spacing value or the intra-group spacing value is selected, and the number of clusters of this clustering is used as the number of categories of the item. The embodiment of the present invention traverses 2-6 clusters, performs clustering respectively, and selects the most appropriate number of clusters (i.e., the number of classifications) by the inter-group spacing value or the intra-group spacing value, which can significantly improve the accuracy of the number of classifications without the need for complex calculations.

[0078] Step 102, calculating the cumulative proportion of the shipment quantity and the cumulative proportion of the number of items of each item according to the historical shipment quantity of each item.

[0079] In this step, the historical outbound volume of each item in a recent period of time (usually at least 1 month) is counted, and then the cumulative proportion of the outbound volume and the cumulative proportion of the number of items of each item are calculated based on the historical outbound volume of each item.

[0080] Optionally, step 102 may include: sorting the items according to their historical outbound quantities, thereby obtaining an outbound quantity ranking of the items; calculating the cumulative outbound quantity proportion of the items according to their historical outbound quantities; and calculating the cumulative proportion of the number of items of the items according to the outbound quantity ranking of the items. After calculating the historical outbound quantity of each item, sort the items in descending order according to the historical outbound quantity from large to small, i.e., the larger the outbound quantity, the higher the ranking; then, the following formula is used to calculate the cumulative proportion of the outbound quantity of each item:

[0081] Cumulative proportion of outbound volume (%) = historical outbound volume of each item / sum of historical outbound volumes of all items * 100%;

[0082] The following formula is used to calculate the cumulative proportion of the number of items for each item:

[0083] Cumulative proportion of item quantity (%) = the order of each item’s shipment quantity / the quantity of all items*100%.

[0084] Step 103, solving the parameters in the regression analysis model based on the cumulative proportion of the shipment quantity of each item and the cumulative proportion of the number of items, so as to obtain a classification model.

[0085] The embodiment of the present invention fits the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items through a regression analysis model, thereby obtaining a classification model, and the classification model can accurately calculate the boundaries corresponding to each classification quantity.

[0086] Optionally, step 103 may include: constructing a regression analysis model; wherein the regression analysis model describes the regression relationship between the cumulative proportion of the outbound quantity and the cumulative proportion of the number of items; constructing an objective function according to the regression analysis model, and solving the objective function based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to minimize the difference between the fitted cumulative proportion of the outbound quantity and the actual cumulative proportion of the outbound quantity; obtaining the parameters in the regression analysis model according to the solution result of the objective function, thereby obtaining a classification model.

[0087] The embodiment of the present invention adopts a very simple and practical regression analysis model:

[0088]

[0089] Among them, y represents the cumulative proportion of the item’s outbound quantity, and x represents the cumulative proportion of the item’s number of items.

[0090] The regression analysis model is close to the distribution characteristics of the outbound quantity of items, so it will have a better fitting effect; and when calculating the parameters of the model, it is only necessary to solve the linear equations, so it is easier to solve and the solution speed is faster.

[0091] Next, the objective function is constructed according to the regression analysis model and solved. The objective function is as follows:

[0092]

[0093] By taking the first-order partial derivatives of the objective function parameters α and β, we can further solve the canonical equations:

[0094]

[0095] Since the above formula is a system of linear equations, α and β can be quickly solved, thereby quickly obtaining a classification model.

[0096] Step 104, based on the number of categories and the classification model, calculate the boundaries of each category, so as to determine the category to which each item belongs.

[0097] In this step, based on the determined classification data and the solved classification model, the boundary points of each classification, that is, the cumulative proportion of the number of items at the boundary points, are calculated, so that the classification of each item is determined based on the cumulative proportion of the number of items at the boundary points and the cumulative proportion of the number of items of each item.

[0098] Optionally, step 104 may include:

[0099] If the number of categories is 2, the cumulative proportion of the number of items at the inflection point of the classification model is calculated, so as to determine the category of each item according to the cumulative proportion of the number of items at the inflection point.

[0100] Specifically, the classification model is actually a curve, so the point with a slope of 1 on the curve can be regarded as the inflection point. Find the inflection point by taking the first derivative and setting it equal to 1.

[0101] At the turning point,

[0102] If the number of categories is 2, that is, divided into two categories, The items are classified into one category. Items are of another category.

[0103] If the number of categories is 3, the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point are solved based on the classification model, so as to determine the category of each item according to the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point.

[0104] If the number of categories is 3, that is, divided into three categories, let the dividing point between category A and category B on the curve be (x1, y1), and the dividing point between category B and category C on the curve be (x2, y2)

[0105] Since the dividing point (x1, y1) and the dividing point (x2, y2) are both On the curve, we have:

[0106]

[0107] In addition, the ratio of the number of items in each category can be determined according to actual needs or experience, so y2-y1=x2-x1=T can be obtained. Among them, T can be determined according to actual needs or experience to determine the ratio of the number of items in each category, for example, it can be within the range of 0.2-0.3.

[0108] Therefore, by combining the above four equations, we can solve the four unknowns, namely x1, y1, x2, y2, so as to determine the boundary points of the three categories. Items with x≤x1 are Class A, items with x1<x≤x2 are Class B, and items with x>x3 are Class C.

[0109] If the number of categories is greater than 3, the items are divided into three categories, and at least one of the three categories is divided equally to determine the category of the items. Optionally, at least one of the three categories is divided equally to determine the category of the items, including: let m be the integer part of k / 3, and n be the remainder of k / 3; where k is the number of categories; if n is equal to 1, the first category is divided equally into m+1 categories, and the second and third categories are divided equally into m categories respectively; if n is equal to 2, the first and second categories are divided equally into m+1 categories respectively, and the third category is divided equally into m categories; the boundary points of each category are determined according to the results of the equal-distance division, so as to determine the category of each item according to the cumulative proportion of the number of items at the boundary points of each category.

[0110] If the number of categories is greater than 3, that is, when there are more than 3 categories, let m be the integer part of k / 3 and n be the remainder, then n=1 or n=2; first, divide all items into three categories according to the classification method when the number of categories is equal to 3, and then subdivide each category equally according to the values ​​of m and n:

[0111] When n=1, class A is divided into m+1 classes at equal intervals, and classes B and C are divided into m classes at equal intervals respectively;

[0112] When n=2, class A and class B are equally divided into m+1 classes, and class C is equally divided into m classes.

[0113] For example, when k=4, m=1, n=1; first, all items are divided into three categories according to the classification method with the number of categories equal to 3, and then category A is equally divided into two subcategories. Since m=1, there is no need to equally divide category B and category C into multiple subcategories.

[0114] The embodiments of the present invention can classify items into any number of categories, not just 2 or 3 categories. On the one hand, the classification method conforms to the basic principle of ABC, that is, category A is the best-selling items, category C is the slow-selling items, and category B is the middle items. Since the slope of the curve of the classification model changes from positive infinity to 0, there must be an area where Δx=Δy, and this area is very suitable for category B. Then, each major category is equally subdivided to achieve the classification of all items.

[0115] Finally, based on the classification results, each item is classified and stored according to the above classification results.

[0116] According to the various embodiments described above, it can be seen that the embodiments of the present invention determine the number of categories according to the historical outbound volume of each item, solve the parameters in the regression analysis model based on the cumulative proportion of the outbound volume of each item and the cumulative proportion of the number of items, thereby obtaining a classification model, and finally calculating the boundaries of each classification based on the number of categories and the classification model. This technical means solves the technical problems in the prior art of subjectively determining the number of categories, classification boundaries, or solving model parameters with high complexity. The embodiments of the present invention can be driven by actual data rather than subjectively determining the number of categories and classification boundaries. The embodiments of the present invention first determine how many categories the items should be divided into based on the historical outbound volume, then quickly build a classification model through a regression analysis model and the historical outbound volume, and quickly calculate the classification boundaries through the classification model. In the embodiments of the present invention, the number of categories can be unlimited, and it is simple and easy to solve, which is convenient for implementation.

[0117] Figure 2 Schematic diagram of the main process of a method for determining the storage classification of items according to a reference embodiment of the present invention. As another embodiment of the present invention, Figure 2 As shown, the method for determining the storage classification of items may include:

[0118] Step 201, obtaining the historical shipment quantity of each item in a recent period of time.

[0119] You can obtain the historical outbound quantity of each item in the last month or the last three months.

[0120] Step 202, calculating the cumulative proportion of the shipment quantity and the cumulative proportion of the number of items of each item according to the historical shipment quantity of each item.

[0121] First, sort the items in descending order according to the historical outbound volume, that is, the larger the outbound volume, the higher the ranking; then, use the following formula to calculate the cumulative proportion of outbound volume of each item:

[0122] Cumulative proportion of outbound volume (%) = historical outbound volume of each item / sum of historical outbound volumes of all items * 100%;

[0123] The following formula is used to calculate the cumulative proportion of the number of items for each item:

[0124] Cumulative proportion of item quantity (%) = the order of each item’s shipment quantity / the quantity of all items*100%.

[0125] The calculation results are shown in the following table:

[0126] SKU Outgoing volume Sorting of outbound quantity Cumulative proportion of outbound volume (%) Cumulative proportion of items (%) 1 200 1 20 0.01 2 100 2 30 0.02 3 50 3 35 0.03 … … … … …

[0127] Step 203: clustering the items according to the historical delivery quantity of each item by using a clustering algorithm, so as to determine the number of categories of each item.

[0128] The classification should be carried out according to the ratio (Δy / Δx) of the difference (Δy) of the cumulative proportion of the outgoing quantity (y) to the difference (Δx) of the cumulative proportion of the number of items (x). However, considering that the difference of the cumulative proportion of the number of items is the same, and the difference Δy of the cumulative proportion of the outgoing quantity is the historical outgoing quantity of the items, the embodiment of the present invention can directly cluster the historical outgoing quantity of each item.

[0129] Optionally, the k-means method can be used for clustering. In k-means, the number of clusters must be specified in advance. Considering that in a warehouse storage scenario, the number of clusters is generally 2 to 6, the number of clusters is pre-selected as 2, 3, 4, 5, and 6. These clusters are clustered separately, and then the inter-group spacing value or intra-group spacing value of each clustering is compared to screen out the smallest inter-group spacing value or intra-group spacing value, and the number of clusters of this clustering is used as the number of categories of the items. The embodiment of the present invention traverses 2-6 clusters, performs clustering respectively, and screens out the most appropriate number of clusters (i.e., the number of categories) by the inter-group spacing value or the intra-group spacing value, which can significantly improve the accuracy of the number of categories without the need for complex calculations.

[0130] Step 204, constructing a regression analysis model; wherein the regression analysis model describes the regression relationship between the cumulative proportion of the shipment volume and the cumulative proportion of the number of items.

[0131] Optionally, the regression analysis model is as follows:

[0132]

[0133] Among them, y represents the cumulative proportion of the outbound quantity of the item, x represents the cumulative proportion of the number of items, and ε i Represents the random error term and can be set in advance.

[0134] The regression analysis model is close to the distribution characteristics of the outbound quantity of items, so it will have a better fitting effect; and when calculating the parameters of the model, it is only necessary to solve the linear equations, so it is easier to solve and the solution speed is faster.

[0135] Step 205, constructing an objective function according to the regression analysis model, and solving the objective function based on the cumulative proportion of the shipment volume of each item and the cumulative proportion of the number of items, so as to minimize the difference between the fitted cumulative proportion of the shipment volume and the actual cumulative proportion of the shipment volume.

[0136] In order to solve the various parameters in the regression analysis model, it is necessary to construct an objective function and solve the objective function, which is as follows:

[0137]

[0138] By taking the first-order partial derivatives of the objective function parameters α and β, we can further solve the canonical equations:

[0139]

[0140] Since the above formula is a system of linear equations, α and β can be quickly solved, thereby quickly obtaining a classification model.

[0141] For example, Figure 3 As shown, there are 10 items, x i The values ​​of y are 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0 respectively; i The values ​​of are 0.700, 0.800, 0.850, 0.871, 0.893, 0.914, 0.936, 0.957, 0.979, 1.000. Therefore, the parameter solution results are: α=0.980021, β=0.03073471.

[0142] Step 206, obtaining parameters in the regression analysis model according to the solution result of the objective function, thereby obtaining a classification model.

[0143] The parameter solution results are α and β, which are the parameters in the regression analysis model. Therefore, the fitted curve (i.e., the classification model) is as follows: Figure 4 shown.

[0144] Step 207, based on the number of categories and the classification model, calculate the boundaries of each category, so as to determine the category to which each item belongs.

[0145] If the number of categories is 2, the cumulative proportion of the number of items at the inflection point of the classification model is calculated, so as to determine the category of each item according to the cumulative proportion of the number of items at the inflection point.

[0146] Specifically, the classification model is actually a curve, so the point with a slope of 1 on the curve can be regarded as the inflection point. Find the inflection point by taking the first derivative and setting it equal to 1.

[0147] At the turning point,

[0148] If the number of categories is 2, that is, divided into two categories, The items are classified into one category. Items are of another category.

[0149] If the number of categories is 3, the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point are solved based on the classification model, so as to determine the category of each item according to the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point.

[0150] If the number of categories is 3, that is, divided into three categories, let the dividing point between category A and category B on the curve be (x1, y1), and the dividing point between category B and category C on the curve be (x2, y2)

[0151] Since the dividing point (x1, y1) and the dividing point (x2, y2) are both On the curve, we have:

[0152]

[0153] In addition, the ratio of the number of items in each category can be determined according to actual needs or experience, so y2-y1=x2-x1=T can be obtained. Among them, T can be determined according to actual needs or experience to determine the ratio of the number of items in each category, for example, it can be within the range of 0.2-0.3.

[0154] Therefore, by combining the above four equations, we can solve the four unknowns, namely x1, y1, x2, y2, so as to determine the boundary points of the three categories. Items with x≤x1 are Class A, items with x1<x≤x2 are Class B, and items with x>x2 are Class C.

[0155] If the number of categories is greater than 3, the items are divided into three categories, and at least one of the three categories is divided equally to determine the category of the items. Optionally, at least one of the three categories is divided equally to determine the category of the items, including: let m be the integer part of k / 3, and n be the remainder of k / 3; where k is the number of categories; if n is equal to 1, the first category is divided equally into m+1 categories, and the second and third categories are divided equally into m categories respectively; if n is equal to 2, the first and second categories are divided equally into m+1 categories respectively, and the third category is divided equally into m categories; the boundary points of each category are determined according to the results of the equal-distance division, so as to determine the category of each item according to the cumulative proportion of the number of items at the boundary points of each category.

[0156] If the number of categories is greater than 3, that is, when there are more than 3 categories, let m be the integer part of k / 3 and n be the remainder, then n=1 or n=2; first, divide all items into three categories according to the classification method when the number of categories is equal to 3, and then subdivide each category equally according to the values ​​of m and n:

[0157] When n=1, class A is divided into m+1 classes at equal intervals, and classes B and C are divided into m classes at equal intervals respectively;

[0158] When n=2, class A and class B are equally divided into m+1 classes, and class C is equally divided into m classes.

[0159] Finally, based on the classification results, each item is classified and stored according to the above classification results.

[0160] In addition, the specific implementation content of the method for determining the storage classification of items in a reference embodiment of the present invention has been described in detail in the method for determining the storage classification of items described above, so the repeated content will not be described here.

[0161] Figure 5 Schematic diagram of main modules of an apparatus for determining storage classification of items according to an embodiment of the present invention. Figure 5 As shown, the device 500 for determining the storage classification of items includes a determination module 501, a calculation module 502, a solution module 503 and a classification module 504; wherein the determination module 501 is used to determine the classification quantity of each item according to the historical outbound quantity of each item; the calculation module 502 is used to calculate the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items according to the historical outbound quantity of each item; the solution module 503 is used to solve the parameters in the regression analysis model based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to obtain a classification model; the classification module 504 is used to calculate the boundaries of each classification based on the classification quantity and the classification model, so as to determine the classification of each item.

[0162] Optionally, the determining module 501 is further configured to:

[0163] A clustering algorithm is used to cluster the items according to their historical outbound quantities, thereby determining the number of categories of the items.

[0164] Optionally, the determining module 501 is further configured to:

[0165] For each clustering, the number of clusters is pre-specified, and the k-means algorithm is used to cluster the items according to the historical outbound volume of each item;

[0166] The inter-group spacing value or the intra-group spacing value of each clustering is compared, and the number of clusters corresponding to the cluster with the smallest inter-group spacing value or the smallest intra-group spacing value is selected as the classification number of each object.

[0167] Optionally, the calculation module 502 is further used for:

[0168] Sorting the items according to their historical shipment quantities, thereby obtaining a shipment quantity ranking of the items;

[0169] Calculate the cumulative proportion of the outbound quantity of each item according to the historical outbound quantity of each item;

[0170] The cumulative proportion of the number of items of each item is calculated according to the order of the outbound quantity of each item.

[0171] Optionally, the solution module 503 is further used for:

[0172] Constructing a regression analysis model; wherein the regression analysis model describes the regression relationship between the cumulative proportion of the outbound volume and the cumulative proportion of the number of items;

[0173] Constructing an objective function according to the regression analysis model, and solving the objective function based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to minimize the difference between the fitted cumulative proportion of the outbound quantity and the actual cumulative proportion of the outbound quantity;

[0174] The parameters in the regression analysis model are obtained according to the solution result of the objective function, thereby obtaining a classification model.

[0175] Optionally, the classification module 504 is further configured to:

[0176] If the number of categories is 2, the cumulative proportion of the number of items at the inflection point of the classification model is calculated, so as to determine the category of each item according to the cumulative proportion of the number of items at the inflection point;

[0177] If the number of categories is 3, the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point are solved based on the classification model, so as to determine the category of each item according to the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point;

[0178] If the number of categories is greater than 3, the items are divided into three categories, and at least one of the three categories is divided equally, so as to determine the category to which the items belong.

[0179] Optionally, the classification module 504 is further configured to:

[0180] Let m be the integer part of k / 3, and n be the remainder of k / 3; where k is the number of categories;

[0181] If n is equal to 1, the first category is divided into m+1 categories with equal intervals, and the second and third categories are divided into m categories with equal intervals; if n is equal to 2, the first and second categories are divided into m+1 categories with equal intervals, and the third category is divided into m categories with equal intervals;

[0182] The boundary points of each category are determined according to the equidistant division results, and the category of each item is determined according to the cumulative proportion of the number of items at the boundary points of each category.

[0183] It should be noted that the specific implementation content of the device for determining the storage classification of items in the present invention has been described in detail in the method for determining the storage classification of items described above, so the repeated content will not be described here.

[0184] Figure 6 An exemplary system architecture 600 is shown to which the method for determining storage classification of items or the apparatus for determining storage classification of items according to an embodiment of the present invention can be applied.

[0185] like Figure 6 As shown, system architecture 600 may include terminal devices 601, 602, 603, network 604 and server 605. Network 604 is used to provide a medium for communication links between terminal devices 601, 602, 603 and server 605. Network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0186] Users can use terminal devices 601, 602, and 603 to interact with server 605 through network 604 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 601, 602, and 603, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0187] The terminal devices 601 , 602 , and 603 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0188] The server 605 may be a server that provides various services, such as a backend management server (for example only) that provides support for shopping websites browsed by users using the terminal devices 601, 602, and 603. The backend management server may analyze and process the received data such as the item information query request, and feed back the processing results to the terminal device.

[0189] It should be noted that the method for determining the storage classification of items provided in the embodiment of the present invention is generally executed by the server 605, and accordingly, the device for determining the storage classification of items is generally set in the server 605. The method for determining the storage classification of items provided in the embodiment of the present invention can also be executed by the terminal devices 601, 602, and 603, and accordingly, the device for determining the storage classification of items can be set in the terminal devices 601, 602, and 603.

[0190] It should be understood that Figure 6 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 implementation requirements.

[0191] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of a terminal device suitable for implementing an embodiment of the present invention. Figure 7 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0192] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0193] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0194] In particular, according to the embodiments disclosed in the present invention, 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 invention include a computer program, 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 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0195] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, 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 invention, 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 invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. 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.

[0196] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program according to various embodiments of the present invention. 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 a 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.

[0197] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes a determination module, a calculation module, a solution module and a classification module, wherein the names of these modules do not constitute a limitation on the modules themselves in some cases.

[0198] As another aspect, the present invention 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 implements the following method: according to the historical outbound quantity of each item, determine the number of categories of the item; according to the historical outbound quantity of each item, calculate the cumulative proportion of the outbound quantity and the cumulative proportion of the number of items of the item; based on the cumulative proportion of the outbound quantity and the cumulative proportion of the number of items of the item, solve the parameters in the regression analysis model, thereby obtaining a classification model; based on the number of categories and the classification model, calculate the boundaries of each category, thereby determining the category of each item.

[0199] As another aspect, an embodiment of the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.

[0200] According to the technical solution of the embodiment of the present invention, because the number of categories is determined according to the historical outbound quantity of each item, the parameters in the regression analysis model are solved based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, thereby obtaining the classification model, and finally the boundaries of each classification are calculated based on the number of categories and the classification model. Therefore, the technical problem of subjectively determining the number of categories, classification boundaries or solving the model parameters with high complexity in the prior art is overcome. The embodiment of the present invention can be driven by actual data instead of subjectively determining the number of categories and classification boundaries. The embodiment of the present invention first determines how many categories the items should be divided into based on the historical outbound quantity, and then quickly constructs the classification model through the regression analysis model and the historical outbound quantity, and quickly calculates the classification boundaries through the classification model. In the embodiment of the present invention, the number of categories can be unlimited, and it is simple and easy to solve, which is convenient for implementation.

[0201] The above specific implementations do not constitute a limitation on the protection scope of the present invention. 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 modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for determining storage classification of items, characterized in that: include: Determine the classification quantity of each item according to the historical outbound quantity of each item; Calculate the cumulative proportion of the shipment quantity and the cumulative proportion of the number of items of each item according to the historical shipment quantity of each item; Solving the parameters in the regression analysis model based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, thereby obtaining a classification model; Based on the number of categories and the classification model, the boundaries of each category are calculated to determine the category to which each item belongs; According to the historical outbound quantity of each item, the classification quantity of each item is determined, including: Clustering the items according to their historical outbound quantities using a clustering algorithm, thereby determining the number of categories for the items; Clustering algorithms are used to cluster the items according to their historical outbound quantities, thereby determining the number of categories of the items, including: For each clustering, the number of clusters is pre-specified, and the k-means algorithm is used to cluster the items according to the historical outbound volume of each item; The inter-group spacing value or the intra-group spacing value of each clustering is compared, and the number of clusters corresponding to the cluster with the largest inter-group spacing value or the smallest intra-group spacing value is selected as the classification number of each object.

2. The method according to claim 1, characterized in that The cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items are calculated according to the historical outbound quantity of each item, including: Sorting the items according to their historical shipment quantities, thereby obtaining a shipment quantity ranking of the items; Calculate the cumulative proportion of the outbound quantity of each item according to the historical outbound quantity of each item; The cumulative proportion of the number of items of each item is calculated according to the order of the outbound quantity of each item.

3. The method according to claim 1, characterized in that The parameters in the regression analysis model are solved based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to obtain a classification model, including: Constructing a regression analysis model; wherein the regression analysis model describes the regression relationship between the cumulative proportion of the outbound volume and the cumulative proportion of the number of items; Constructing an objective function according to the regression analysis model, and solving the objective function based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to minimize the difference between the fitted cumulative proportion of the outbound quantity and the actual cumulative proportion of the outbound quantity; The parameters in the regression analysis model are obtained according to the solution result of the objective function, thereby obtaining a classification model.

4. The method according to claim 1, characterized in that: Based on the number of categories and the classification model, the boundaries of each category are calculated to determine the category to which each item belongs, including: If the number of categories is 2, the cumulative proportion of the number of items at the inflection point of the classification model is calculated, so as to determine the category of each item according to the cumulative proportion of the number of items at the inflection point; If the number of categories is 3, the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point are solved based on the classification model, so as to determine the category of each item according to the cumulative proportion of the number of items at the first boundary point and the cumulative proportion of the number of items at the second boundary point; If the number of categories is greater than 3, the items are divided into three categories, and at least one of the three categories is divided equally, so as to determine the category to which the items belong.

5. The method according to claim 4, characterized in that At least one of the three categories is divided into equal intervals to determine the category of each item, including: Let m be the integer part of k / 3, and n be the remainder of k / 3; where k is the number of categories; If n is equal to 1, the first category is divided into m+1 categories with equal intervals, and the second and third categories are divided into m categories with equal intervals; if n is equal to 2, the first and second categories are divided into m+1 categories with equal intervals, and the third category is divided into m categories with equal intervals; The boundary points of each category are determined according to the equidistant division results, and the category of each item is determined according to the cumulative proportion of the number of items at the boundary points of each category.

6. A device for determining the storage classification of items, characterized in that: include: A determination module, used to determine the classification quantity of each item according to the historical outbound quantity of each item; A calculation module, used to calculate the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items according to the historical outbound quantity of each item; A solution module, used to solve the parameters in the regression analysis model based on the cumulative proportion of the outbound quantity of each item and the cumulative proportion of the number of items, so as to obtain a classification model; A classification module, used to calculate the boundaries of each classification based on the number of classifications and the classification model, so as to determine the classification of each item; The determining module is also used for: Clustering the items according to their historical outbound quantities using a clustering algorithm, thereby determining the number of categories for the items; The determining module is also used for: For each clustering, the number of clusters is pre-specified, and the k-means algorithm is used to cluster the items according to the historical outbound volume of each item; The inter-group spacing value or the intra-group spacing value of each clustering is compared, and the number of clusters corresponding to the cluster with the largest inter-group spacing value or the smallest intra-group spacing value is selected as the classification number of each object.

7. An electronic device, 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 5.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.