CN111738657B - Logistics bill splitting optimization method and system based on substitute recommendation - Google Patents
Logistics bill splitting optimization method and system based on substitute recommendation Download PDFInfo
- Publication number
- CN111738657B CN111738657B CN202010568808.5A CN202010568808A CN111738657B CN 111738657 B CN111738657 B CN 111738657B CN 202010568808 A CN202010568808 A CN 202010568808A CN 111738657 B CN111738657 B CN 111738657B
- Authority
- CN
- China
- Prior art keywords
- user
- commodity
- substitute
- commodities
- purchased
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The invention provides a method and a system for optimizing logistics splitting list based on substitute recommendation, and relates to the technical field of data processing. The method comprises the steps of firstly, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located; then, based on historical scoring data, the utility of the user to the commodity without scoring data is obtained; obtaining a maximum profit objective function and constraints based on the user's utility and other data for the good without the scoring data; and finally, acquiring a commodity recommendation result based on the maximum profit objective function and the constraint condition. According to the invention, warehouse information at the rear end and ordering behaviors of consumers are considered, and an individualized recommendation technology is combined, so that the consumers are guided to change purchasing behaviors from the preference of the consumers to commodities, and the same-bin substitute combination is recommended for the consumers, thereby realizing the reduction of the order splitting rate.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a system for optimizing logistics splitting list based on substitute recommendation.
Background
With the rapid development of electronic commerce, the convenience of online shopping attracts more and more consumers to shop online. Many commodities are not stored in a single warehouse due to limited capacity of the warehouse and cost-based considerations. A large number of orders can be split into sub-orders for separate distribution, logistics cost is increased, and shopping experience of consumers is reduced due to the fact that goods taking times are increased. In fact, for enterprises, the marginal cost of additional logistics caused by adding one commodity in a package is very low, so if as many commodities as possible are put in one package, the number of detached lists can be reduced, the logistics cost of the enterprises can be remarkably saved, and the shopping experience of consumers is improved.
The existing method for reducing the order splitting rate is mainly optimized from the aspects of distribution sequence and classification distribution of back-end design orders, centralized placement of hot-sold commodities and associated commodities and the like, so that the purposes of reducing the order splitting rate and reducing the logistics cost are achieved.
However, no researcher has been found to optimize the existing warehouse information at the back end and the ordering behavior of the consumer, so as to reduce the order splitting rate.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a logistics form splitting optimization method and a logistics form splitting optimization system based on substitute recommendation, which are used for guiding a consumer to change the original purchase combination by utilizing a personalized recommendation technology based on the existing warehouse information at the rear end and the order placing behavior of the consumer, thereby realizing the reduction of the form splitting rate.
(II) technical scheme
In order to realize the purpose, the invention is realized by the following technical scheme:
the invention provides a logistics bill splitting optimization method based on substitute recommendation, which comprises the following steps:
s1, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located;
s2, obtaining the utility of the user to the commodity without scoring data based on the historical scoring data;
s3, obtaining a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost;
and S4, obtaining a commodity recommendation result based on the maximum profit objective function and the constraint condition.
Preferably, the utility of the user to the goods without score data based on the historical score data includes:
s201, constructing a user-commodity utility matrix based on the historical scoring data;
s202, acquiring a feature vector of a user and a feature vector of a commodity based on a user-commodity utility matrix;
s203, obtaining the utility of the user to the commodity without the scoring data based on the point multiplication measurement between the feature vector of the user and the feature vector of the commodity.
Preferably, the constructing a user-commodity utility matrix based on the historical scoring data comprises:
representing user information data in historical score data as P = { P = 1 ,p 2 ,...,p u ,...,p n In which p is u Represents the u-th user;
expressing the commodity information data in the history score data as S = { S = } 1 ,s 2 ,...,s i ,...,s m In which s i Indicates the ith product;
the score of the ith user on the ith commodity is n ui Using the score as a measure of the user's utility for the item;
establishing a user-commodity utility matrix X epsilon R n×m Where each row represents a user's rating of all items, each column represents a generationTable a rating of an item by all users.
Preferably, the obtaining the utility of the user for the commodity without the score data based on the point multiplication metric between the feature vector of the user and the feature vector of the commodity comprises:
and (3) scoring the commodity i by the user u, modeling through point multiplication between the user characteristic vector and the commodity characteristic vector, wherein the objective function is as follows:
wherein:
λ is a regularization coefficient;
||·|| 2 represents the L2 regularization term;
P∈R k×n a feature matrix representing users, each column representing a feature vector of a user, S ∈ R k×m Representing a feature matrix of the commodity, wherein each column represents a feature vector of the commodity, and k is a set dimension number;
by minimizing the objective function, learning the user characteristic matrix P ∈ R k×n And the commodity feature matrix S ∈ R k×m ；
Obtaining user utility of goods without scoring data through point multiplication
Preferably, the obtaining of the maximum profit objective function and the constraint condition based on the utility of the user to the commodity without the score data, the set of commodities the user wants to purchase, the warehouse information of the commodities the user wants to purchase, the price information of the substitute of the warehouse where the commodities the user wants to purchase, and the logistics cost includes:
the method comprises the steps that users intend to purchase two commodities i and j, a warehouse comprises a local warehouse and a total warehouse, the local warehouse comprises a% of total commodities, the total warehouse comprises b% of the total commodities, and the 0-straw-a-straw-b-straw 100; analyzing the commodities to be purchased by the user, wherein order splitting optimization is not needed if the two commodities are in the same warehouse, or the order splitting optimization is performed if the two commodities are not in the same warehouse, or a substitute of one of the commodities is needed if the two commodities are not in the same warehouse; comparing the expected income of the new combination with the income of the original combination, and taking the maximum combination;
the maximum profit objective function and constraints are expressed as follows:
s.t.
V u ＝r i +r j -(c i +c j ) (2)
X ij' +X i'j ＝1 (4)
wherein:
u represents a set of users;
s represents a set of commodities;
p i represents the price of item i;
q i represents the cost of commodity i;
r i indicates the profit of the item i, r i ＝p i -q i
c i Represents the logistics cost of the commodity i, c i' Represents the logistics cost of the substitute i', c j Denotes the logistics cost of the item j, c j' Represents the logistics cost of substitute j';
i represents a substitute set of a commodity I, I belongs to I;
j represents a substitute set of a commodity J, wherein J belongs to J;
i 'represents a substitute of a commodity I, I' belongs to I;
j 'represents a substitute of commodity J, and J' belongs to J;
n ui indicating the utility of user u for item i,
indicating the level of utility of the substitute relative to the original commodity, the decline in utility can be compensated for by a price discount;
p ui' represents a preferential price for the user u to purchase a substitute item i' for the item i,
λ represents an extra discount coefficient representing the sharing of extra profit generated by reducing the logistics cost due to bill breaking optimization, and when λ =1, represents that the user u does not participate in the sharing of extra profit generated by reducing the logistics cost, 0<λ≤1；
X ij' indicates whether a substitute ij' combination, X, is recommended ij' =1 recommendation, X ij' =0 not recommended;
X i'j indicates whether a substitute i' j combination, X, is recommended i'j =1 recommendation, X i'j =0 not recommended;
constraint (1) indicates that the recommended substitute is the one with the largest current expected benefit;
the constraint (2) represents the expected profit margin of the consumer when purchasing the original goods;
the constraint condition (3) indicates that the price after the commodity is discounted is not lower than the cost price and does not exceed the price equal to the effectiveness of the original commodity, namely the recommended commodity cannot be sold at the loss;
constraint (4) indicates that only one substitute is recommended to the consumer while only one of the two commodities can be selected.
Preferably, the obtaining of the recommendation result of the goods based on the maximum profit objective function and the constraint condition includes:
s401, selecting a user U belonging to U from a user set, and acquiring commodities i and j to be purchased by the user;
s402, judging whether the commodities i and j are in the same bin, and if the commodities i and j are in different bins, executing the next step; otherwise, entering step 1;
s403, searching a substitute collection J of the commodity J from a warehouse where the commodity i is located;
s404, selecting a substitute J 'from the set J to form an ij' substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, the corresponding purchase probability is obtained, the expected income at the moment is calculated, the lambda at the moment with the maximum expected income is selected, and the expected income v of the enterprise of the substitute combination ij' at the moment is calculated ij' ；
S405, repeating the step S404 until all commodities in the set J are traversed, selecting the substitute combination with the maximum expected income, and using V for the expected income ij' Represents;
s406, selecting a substitute I 'from the set I to form an I' j substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, the corresponding purchase probability is obtained, the expected income at the moment is calculated, the lambda at the moment with the maximum expected income is selected, and the expected income v of the enterprise with the substitute combination i' j at the moment is calculated i'j ；
S407, repeatedly executing the step S406 until all commodities in the set I are traversed, selecting the substitute combination with the maximum expected income, wherein the expected income is V i'j Represents;
s408, calculating the income V when the original combination ij is purchased ij ；
S409, comparison V ij' 、V i'j And V ij Selecting a combination with the highest expected profit to recommend to the consumer.
Preferably, the method further comprises:
and repeatedly executing the steps S401 to S409 until all the users in the set U are traversed, and calculating the final bill splitting rate and the final enterprise profit increase rate.
The invention also provides a logistics list splitting optimization system based on substitute recommendation, which comprises a computer, wherein the computer comprises:
at least one memory cell;
at least one processing unit;
wherein the at least one memory unit has stored therein at least one instruction that is loaded and executed by the at least one processing unit to perform the steps of:
s1, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located;
s2, obtaining the utility of the user to the commodity without scoring data based on the historical scoring data;
s3, obtaining a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost;
and S4, obtaining a commodity recommendation result based on the maximum profit objective function and the constraint condition.
(III) advantageous effects
The invention provides a method and a system for optimizing logistics bill splitting based on substitute recommendation. Compared with the prior art, the method has the following beneficial effects:
the method comprises the steps of firstly, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located; then, based on historical scoring data, the utility of the user to the commodity without scoring data is obtained; acquiring a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost; and finally, acquiring a commodity recommendation result based on the maximum profit objective function and the constraint condition. According to the invention, the warehouse information at the back end and the ordering behavior of the consumer are considered, and the personalized recommendation technology is combined, so that the consumer is guided to change the purchasing behavior from the preference of the consumer to the commodity, and the same-bin substitute combination is recommended for the consumer, thereby realizing the reduction of the order-dismantling rate.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a block diagram of a logistics de-ordering optimization method based on substitute recommendation in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described, and it is obvious that the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the application provides a logistics list splitting optimization method based on substitute recommendation, solves the problem that optimization is not performed in the prior art based on the existing warehouse information at the rear end and the list dropping behavior of a consumer, and achieves reduction of list splitting rate.
In order to solve the technical problems, the general idea of the embodiment of the present application is as follows:
splitting the order tends to increase the order fulfillment cost for the business, as well as reduce the shopping experience for the consumer. The conventional research is generally optimized from the back end, namely the middle links from the order placement to the delivery to the hands of the consumer (the goods taking point), the embodiment of the invention provides a logistics bill splitting optimization method based on substitute recommendation by utilizing a personalized recommendation technology based on the conventional warehouse information at the back end and the order placement behavior of the consumer, and the reduction of the bill splitting rate is realized by guiding the consumer to change the original purchasing combination and recommending the same-warehouse substitute combination for the consumer.
In order to better understand the technical solution, the technical solution will be described in detail with reference to the drawings and the specific embodiments.
The embodiment of the invention provides a logistics list splitting optimization method based on substitute recommendation, which comprises the following steps of S1-S4:
s1, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located;
s2, obtaining the utility of the user to the commodity without scoring data based on historical scoring data;
s3, obtaining a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost;
and S4, acquiring a commodity recommendation result based on the maximum profit objective function and the constraint condition.
According to the embodiment of the invention, warehouse information at the rear end and ordering behaviors of consumers are considered, and personalized recommendation technology is combined, so that the consumers are guided to change purchasing behaviors from the preference of the consumers to commodities, and the same-bin substitute combination is recommended for the consumers, thereby realizing the reduction of the order splitting rate.
Each step is described in detail below.
In step S1, historical rating data of the user on the commodity, a set of commodities the user wants to purchase, warehouse information of the commodities the user wants to purchase, and price information and logistics cost of a substitute of a warehouse where the commodities the user wants to purchase are obtained. The method specifically comprises the following steps:
the historical grading data of the user on the commodities, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of substitutes of the warehouse where the commodities to be purchased by the user are located, the logistics cost and the like are obtained through manual input or a crawler technology.
In step S2, the utility of the user for the goods without rating data is acquired based on the historical rating data. The method specifically comprises the following steps:
s201, constructing a user-commodity utility matrix based on the historical scoring data. The specific implementation process is as follows:
representing user information data in historical score data as P = { P = 1 ,p 2 ,...,p u ,...,p n In which p is u Represents the u-th user;
expressing the commodity information data in the history score data as S = { S = } 1 ,s 2 ,...,s i ,...,s m In which s i Represents the ith product;
the score of the ith user on the ith commodity is n ui Using the score as a measure of the user's utility for the item;
establishing a two-dimensional sparse utility matrix X belonging to R of a user-commodity n×m Wherein each row represents the rating of all items by one user and each column represents the rating of all items by one user.
S202, acquiring a characteristic vector of the user and a characteristic vector of the commodity based on the user-commodity utility matrix. The specific implementation process is as follows:
the utility matrix of user-goods can be expressed as X = P T S, wherein P ∈ R k×n A feature matrix representing users, each column representing a feature vector of a user, S ∈ R k×m And each column represents a feature vector of the commodity, wherein k is the set dimension number.
S203, obtaining the utility of the user to the commodity without the scoring data based on the point multiplication measurement between the feature vector of the user and the feature vector of the commodity. The specific implementation process is as follows:
the scoring of the commodity i by the user u can be modeled by point multiplication between the user feature vector and the commodity feature vector.
The objective function is:
wherein:
λ is a regularization coefficient;
||·|| 2 representing the L2 regularization term.
Learning a user feature matrix P epsilon R by minimizing an objective function k×n And the item feature matrix S ∈ R k×m And overfitting is avoided by the regularization term.
Obtaining the grade of the user to the commodity without grade data through point multiplication
This score is used as a measure of the utility of user u for item i.
It should be noted that, in the embodiment of the present invention, the above method is only one method for obtaining the utility of the user on the commodity without the score data, and in the technical field of personalized recommendation, various methods are included to obtain the utility of the user on the commodity without the score data, such as a collaborative filtering recommendation technology based on the user and a collaborative filtering recommendation technology based on the product.
In step S3, a maximum profit objective function and a constraint condition are obtained based on the utility of the user for the commodity without the score data, the set of commodities the user wants to purchase, the warehouse information of the commodities the user wants to purchase, and the price information and the logistics cost of the substitute of the warehouse where the commodities the user wants to purchase. The method specifically comprises the following steps:
the method comprises the steps that users intend to purchase two commodities i and j, a warehouse comprises a local warehouse and a total warehouse, the local warehouse comprises a% of total commodities, the total warehouse comprises b% of the total commodities, and the 0-straw-a-straw-b-straw 100; analyzing the commodities to be purchased by the user, wherein order splitting optimization is not needed if the two commodities are in the same warehouse, or the order splitting optimization is performed if the two commodities are not in the same warehouse, or a substitute of one of the commodities is needed if the two commodities are not in the same warehouse; comparing the expected income of the new combination with the income of the original combination, and taking the maximum combination;
the maximum profit objective function and constraints are expressed as follows:
s.t.
V u ＝r i +r j -(c i +c j )(2)
X ij' +X i'j ＝1 (4)
wherein:
u represents a set of users;
s represents a set of commodities;
p i represents the price of item i;
q i represents the cost of commodity i;
r i indicates the profit of the commodity i, r i ＝p i -q i
c i Represents the logistics cost of the commodity i, c i' Represents the logistics cost of the substitute i', c j Denotes the logistics cost of the item j, c j' Represents the logistics cost of substitute j';
i represents a substitute set of a commodity I, I belongs to I;
j represents a substitute set of a commodity J, and J belongs to J;
i 'represents a substitute of a commodity I, I' belongs to I;
j 'represents a substitute of a commodity J, and J' belongs to the group J;
n ui indicating the utility of user u for item i,
indicating the level of utility of the substitute relative to the original commodity, the decline in utility can be compensated for by a price discount;
p ui' represents a preferential price for the user u to purchase a substitute item i' for the item i,
λ represents an extra discount coefficient, which represents sharing of extra revenue generated by reduction of logistics cost due to bill splitting optimization, and when λ =1, which represents that user u does not participate in sharing of extra revenue generated by reduction of logistics cost, 0<λ≤1；
X ij' indicates whether a substitute ij' combination, X, is recommended ij' =1 recommendation, X ij' =0 not recommended;
X i'j indicates whether a substitute i' j combination, X, is recommended i'j =1 recommendation, X i'j =0 not recommended;
the constraint (1) indicates that the recommended substitute is the one with the largest current expected profit;
the constraint (2) represents the expected profit margin of the consumer when purchasing the original commodity;
the constraint condition (3) indicates that the price after the commodity is discounted is not lower than the cost price and does not exceed the price equal to the effectiveness of the original commodity, namely the recommended commodity cannot be sold at the loss;
constraint (4) indicates that only one substitute is recommended to the consumer while only one of the two commodities can be selected.
In step S4, a product recommendation is obtained based on the maximum profit objective function and constraint conditions. The method specifically comprises the following steps:
s401, selecting a user U belonging to U from a user set, and acquiring commodities i and j to be purchased by the user;
s402, judging whether the commodities i and j are in the same bin, and if the commodities i and j are in different bins, executing the next step; otherwise, entering step 1;
s403, searching a substitute collection J of the commodity J from a warehouse where the commodity i is located;
s404, selecting a substitute J 'from the set J to form an ij' substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, corresponding purchase probability is obtained, expected income at the moment is calculated, lambda at the moment when the expected income is maximum is selected, and expected income v of the substitute combination ij' enterprise at the moment is calculated ij' ；
S405, repeating the step S404 until all commodities in the set J are traversed, selecting the substitute combination with the maximum expected income, and using V for the expected income ij' Represents;
s406, selecting a substitute I 'from the set I to form an I' j substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, the corresponding purchase probability is obtained, the expected income at the moment is calculated, the lambda at the moment with the maximum expected income is selected, and the expected income v of the enterprise with the substitute combination i' j at the moment is calculated i'j ；
S407, repeatedly executing the step S406 until all commodities in the set I are traversed, selecting the substitute combination with the maximum expected income, wherein the expected income is V i'j Represents;
s408, calculating the income V when the original combination ij is purchased ij ；
S409, comparison V ij' 、V i'j And V ij Selecting a combination with the highest expected profit to recommend to the consumer.
In order to maximize the cooperative income of the consumers and the enterprises in the specific implementation process, the embodiment of the invention further comprises the following steps:
and repeatedly executing the steps S401 to S409 until all the users in the set U are traversed, and calculating the final bill splitting rate and the final enterprise profit increase rate.
In order to verify the effectiveness of the embodiment of the present invention, the following experimental demonstration is performed, specifically including:
suppose that a user wants to purchase two kinds of commodities, the warehouses only have two warehouses, namely a local warehouse and a total warehouse, and the freight rates are respectively 5 yuan and 8 yuan, wherein the local warehouse stores 55% of commodity categories, the total warehouse stores 85% of SKUs (Stock keeping units), 15% of SKUs are unique to the local warehouse, and the profit margin of the commodities is 20%.
1. Data set
An example dataset of an embodiment of the invention is a public dataset from amazon that contains information such as a user's rating of a good, a category of the good, a price of the good, and the like. Preprocessing the data, and finally selecting 214392 users and 100 commodities, wherein 62123 users need to perform order splitting optimization, and the order splitting rate at the moment is 28.98%. And performing dot multiplication on the obtained characteristic vector of the user and the characteristic vector of the commodity through matrix decomposition according to the grade of the commodity by the user to obtain the utility of the user to the commodity not purchased. And sequencing the effectiveness from large to small, and selecting the first two commodities as commodities to be purchased by the user. When the additional discount coefficient λ =1, the utility of the substitute is equal to the original utility of the commodity to be purchased by the consumer, and the user does not participate in the sharing of the additional benefit generated by the cost reduction of the logistics. At which point the user has a probability of purchasing. When the additional discount coefficient lambda>When 1, since the utility of the alternative is lower than that of the original commodity, the probability of the user purchasing the alternative is 0. When the additional discount coefficient λ =0, the probability of the user selecting the substitute is 1. Thus, the additional discount coefficient λ is obtained in relation to the probability of purchasing a substitute
Alpha and beta are hyper-parameters, alpha influences the steepness of the function curve, and beta influences the initial probability value of the function. The values of α and β can be obtained from empirical values in practical application scenarios.
In this experiment, the purchase probability function of the consumer is set as
When the discount rate is reduced, the probability of the user selecting a substitute increases. The desired gain is maximized by adjusting λ. To pairThe substitutes for both commodities are combined, and the combination with the highest expected profit is selected.
2. Results of the experiment
In order to verify the effectiveness of the method provided by the embodiment of the invention, the method is used for testing on amazon commodity data sets, the results of the embodiment of the invention are compared with the results obtained when no substitute recommendation is adopted, the results are shown in tables 1 and 2, the experimental results show that the bill removal rate is reduced to 19.98% after the substitute recommendation is adopted, the bill removal rate is reduced by 9%, and the profit is improved by 4.63%. Therefore, the method provided by the embodiment of the invention can effectively reduce the bill splitting rate.
TABLE 1 Change freight rate of warehouse, get the relationship between freight rate and rate of splitting order and profit increase
Table 2 changes the gross profit rate of the commodity to obtain the relationship between the gross profit rate and the rate of removing bills and increasing profit
The invention also provides a logistics bill splitting optimization system based on substitute recommendation, which comprises a computer, wherein the computer comprises:
at least one memory cell;
at least one processing unit;
wherein, at least one instruction is stored in the at least one storage unit, and the at least one instruction is loaded and executed by the at least one processing unit to realize the following steps:
s1, obtaining historical scoring data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located;
s2, obtaining the utility of the user to the commodity without the grading data based on the historical grading data;
s3, obtaining a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost;
and S4, acquiring a commodity recommendation result based on the maximum profit objective function and the constraint condition.
It can be understood that, the logistics list splitting optimization system based on substitute recommendation provided in the embodiment of the present invention corresponds to the logistics list splitting optimization method based on substitute recommendation, and relevant explanations, examples, and beneficial effects of the relevant contents thereof may refer to corresponding contents in the logistics list splitting optimization method based on substitute recommendation, which are not described herein again.
In summary, compared with the prior art, the method has the following beneficial effects:
1. according to the embodiment of the invention, the warehouse information at the back end and the ordering behavior of the consumer are considered, and the personalized recommendation technology is combined, so that the consumer is guided to change the purchasing behavior from the preference of the consumer to the commodity, and the same-bin substitute combination is recommended for the consumer, thereby realizing the reduction of the order splitting rate.
2. According to the embodiment of the invention, by means of an individualized recommendation technology and in combination with warehouse information, a front-end and back-end combined optimization strategy is provided, so that the order splitting rate can be effectively reduced, the pressure of optimizing a back-end warehouse can be reduced, and the saved logistics cost can be shared by an enterprise and a user together, so that the cooperative income of the consumer and the enterprise is maximized, the shopping experience of the consumer is considered, and the profit maximization of the enterprise is realized.
It should be noted that, through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (6)
1. A logistics bill splitting optimization method based on substitute recommendation is characterized by comprising the following steps:
s1, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located;
s2, obtaining the utility of the user to the commodity without the grading data based on the historical grading data, and the method comprises the following steps:
s201, constructing a user-commodity utility matrix based on the historical scoring data;
s202, acquiring a feature vector of a user and a feature vector of a commodity based on a user-commodity utility matrix;
s203, obtaining the utility of the user to the commodity without the grading data based on the point multiplication measurement between the feature vector of the user and the feature vector of the commodity;
s3, obtaining a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost;
s4, obtaining a commodity recommendation result based on the maximum profit objective function and the constraint condition, wherein the commodity recommendation result comprises the following steps:
s401, selecting a user U belonging to U from a set of users, and acquiring commodities i and j to be purchased by the user, wherein the U represents the set of the users;
s402, judging whether the commodities i and j are in the same bin, and if the commodities i and j are in different bins, executing the next step; otherwise, entering step 1;
s403, searching a substitute set J of the commodity J from a warehouse where the commodity i is located;
s404, selecting one substitute J 'from the set J to form an ij' substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, the corresponding purchase probability is obtained, the expected income at the moment is calculated, the lambda at the moment with the maximum expected income is selected, and the expected income v of the enterprise of the substitute combination ij' at the moment is calculated ij' ；
S405, repeating the step S404 until all commodities in the set J are traversed, selecting the substitute combination with the maximum expected income, and using V for the expected income ij' Represents;
s406, selecting a substitute I 'from the set I to form an I' j substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, corresponding purchase probability is obtained, expected income at the moment is calculated, and expected income is selectedLambda when the profit is maximum, and calculating the expected income v of the enterprise of the substitute combination i' j at the moment i'j ；
S407, repeatedly executing the step S406 until all the commodities in the set I are traversed, selecting the substitute combination with the largest expected income, wherein the expected income is V i'j Represents;
s408, calculating the income V when the original combination ij is purchased ij ；
S409, comparison V ij' 、V i'j And V ij Selecting a combination with the highest expected profit to recommend to the consumer.
2. The method of claim 1, wherein the constructing a user-commodity utility matrix based on the historical scoring data comprises:
representing user information data in historical scoring data as P = { P 1 ,p 2 ,...,p u ,...,p n In which p is u Represents the u-th user;
expressing the commodity information data in the history score data as S = { S = } 1 ,s 2 ,...,s i ,...,s m In which s i Represents the ith product;
the score of the ith user on the ith commodity is n ui Using the score as a measure of the user's utility for the item;
establishing a user-commodity utility matrix X epsilon R n×m Wherein each row represents the rating of all items by one user and each column represents the rating of all items by one user.
3. The method for optimizing logistics breakdown based on substitute recommendation as set forth in claim 2, wherein the obtaining utility of the user for the commodity without score data based on the point-product metric between the feature vector of the user and the feature vector of the commodity comprises:
and (3) scoring the commodity i by the user u, modeling by point multiplication between the user characteristic vector and the commodity characteristic vector, wherein the objective function is as follows:
wherein:
λ is a regularization coefficient;
||·|| 2 represents the L2 regularization term;
P∈R k×n a feature matrix representing users, each column representing a feature vector of a user, S ∈ R k×m Representing a feature matrix of the commodity, wherein each column represents a feature vector of the commodity, and k is a set dimension number;
learning a user feature matrix P epsilon R by minimizing an objective function k×n And the commodity feature matrix S ∈ R k×m ；
Obtaining utility of user to commodity without scoring data through point multiplication
4. The method as claimed in claim 1, wherein the obtaining of the maximum profit objective function and the constraint condition based on the utility of the user to the goods without score data, the collection of the goods to be purchased, the warehouse information of the goods to be purchased, the price information of the substitute of the warehouse where the goods to be purchased are located, and the logistics cost comprises:
the method comprises the steps that users intend to purchase two commodities i and j, a warehouse comprises a local warehouse and a total warehouse, the local warehouse comprises a% of total commodities, the total warehouse comprises b% of the total commodities, and the 0-straw-a-straw-b-straw 100; analyzing the commodities to be purchased by the user, wherein order splitting optimization is not needed if the two commodities are in the same warehouse, or the order splitting optimization is performed if the two commodities are not in the same warehouse, or a substitute of one of the commodities is needed if the two commodities are not in the same warehouse; comparing the expected income of the new combination with the income of the original combination, and taking the maximum combination;
the maximum profit objective function and constraints are expressed as follows:
total profit
V u ＝r i +r j -(c i +c j ) (2)
X ij' +X i'j ＝1 (4)
Wherein:
u represents a set of users;
s represents a set of commodities;
p i represents the price of item i;
q i represents the cost of commodity i;
r i indicates the profit of the item i, r i ＝p i -q i
c i Represents the logistics cost of the item i, c i' Represents the logistics cost of the substitute i', c j Represents the logistics cost of the commodity j, c j' Represents the logistics cost of substitute j';
i represents a substitute set of a commodity I, I belongs to I;
j represents a substitute set of a commodity J, and J belongs to J;
i 'represents a substitute of a commodity I, I' belongs to I;
j 'represents a substitute of commodity J, and J' belongs to J;
n ui indicating the utility of user u for item i,
indicating the level of utility of the substitute relative to the original commodity, the decline in utility being obtainable by a price discountTo compensate;
p ui' represents a preferential price for the user u to purchase a substitute item i' for the item i,
λ represents an extra discount coefficient, which represents sharing of extra revenue generated by reduction of logistics cost due to bill splitting optimization, and when λ =1, which represents that user u does not participate in sharing of extra revenue generated by reduction of logistics cost, 0<λ≤1；
X ij' indicates whether a substitute ij' combination, X, is recommended ij' =1 recommendation, X ij' =0 not recommended;
X i'j indicates whether a substitute i' j combination, X, is recommended i'j =1 recommendation, X i'j =0 not recommended;
constraint (1) indicates that the recommended substitute is the one with the largest current expected benefit;
the constraint (2) represents the expected profit margin of the consumer when purchasing the original commodity;
the constraint condition (3) indicates that the price after the commodity is discounted is not lower than the cost price and does not exceed the price equal to the effectiveness of the original commodity, namely the recommended commodity cannot be sold at the loss;
constraint (4) indicates that only one substitute is recommended to the consumer while only one substitute of the two goods can be selected.
5. The method of claim 1 for logistics de-ordering optimization based on surrogate recommendation, the method further comprising:
and repeatedly executing the steps S401 to S409 until all the users in the set U are traversed, and calculating the final bill splitting rate and the final enterprise profit increase rate.
6. A logistics de-ordering optimization system based on substitute recommendation, the system comprising a computer, the computer comprising:
at least one memory cell;
at least one processing unit;
wherein the at least one memory unit has stored therein at least one instruction that is loaded and executed by the at least one processing unit to perform the steps of:
s1, acquiring historical grading data of a user on commodities, a set of commodities to be purchased by the user, warehouse information of the commodities to be purchased by the user, and price information and logistics cost of substitutes of a warehouse where the commodities to be purchased by the user are located;
s2, obtaining the utility of the user to the commodity without the grading data based on the historical grading data, and the method comprises the following steps:
s201, constructing a user-commodity utility matrix based on the historical scoring data;
s202, acquiring a feature vector of a user and a feature vector of a commodity based on a user-commodity utility matrix;
s203, obtaining the utility of the user to the commodity without the grading data based on the point multiplication measurement between the feature vector of the user and the feature vector of the commodity;
s3, obtaining a maximum profit objective function and constraint conditions based on the utility of the user to the commodity without the grading data, the set of the commodities to be purchased by the user, the warehouse information of the commodities to be purchased by the user, the price information of the substitute of the warehouse where the commodities to be purchased by the user are located, and the logistics cost;
s4, obtaining a commodity recommendation result based on the maximum profit objective function and the constraint condition, wherein the commodity recommendation result comprises the following steps:
s401, selecting a user U belonging to U from a set of users, and acquiring commodities i and j to be purchased by the user, wherein the U represents the set of the users;
s402, judging whether the commodities i and j are in the same bin, and if the commodities i and j are in different bins, executing the next step; otherwise, entering step 1;
s403, searching a substitute set J of the commodity J from a warehouse where the commodity i is located;
s404, selecting one substitute J 'from the set J to form an ij' substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, the corresponding purchase probability is obtained, the expected income at the moment is calculated, the lambda at the moment with the maximum expected income is selected, and the expected income v of the enterprise of the substitute combination ij' at the moment is calculated ij' ；
S405, repeating the step S404 until all commodities in the set J are traversed, selecting the substitute combination with the maximum expected income, and using V for the expected income ij' Represents;
s406, selecting a substitute I 'from the set I to form an I' j substitute combination, calculating the expected income of the combined enterprise, and adjusting the additional discount coefficient lambda, 0<Lambda is less than or equal to 1, the corresponding purchase probability is obtained, the expected income at the moment is calculated, the lambda at the moment with the maximum expected income is selected, and the expected income v of the enterprise with the substitute combination i' j at the moment is calculated i'j ；
S407, repeatedly executing the step S406 until all commodities in the set I are traversed, selecting the substitute combination with the maximum expected income, wherein the expected income is V i'j Representing;
s408, calculating the income V when the original combination ij is purchased ij ；
S409, comparison V ij' 、V i'j And V ij Selecting a combination with the highest expected profit to recommend to the consumer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010568808.5A CN111738657B (en) | 2020-06-19 | 2020-06-19 | Logistics bill splitting optimization method and system based on substitute recommendation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010568808.5A CN111738657B (en) | 2020-06-19 | 2020-06-19 | Logistics bill splitting optimization method and system based on substitute recommendation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111738657A CN111738657A (en) | 2020-10-02 |
| CN111738657B true CN111738657B (en) | 2022-11-04 |
Family
ID=72651871
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010568808.5A Active CN111738657B (en) | 2020-06-19 | 2020-06-19 | Logistics bill splitting optimization method and system based on substitute recommendation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111738657B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679516A (en) * | 2012-09-25 | 2014-03-26 | 付艳超 | Price comparison system based on shopping cart and method thereof |
| CN105913301A (en) * | 2016-04-08 | 2016-08-31 | 珠海优特电力科技股份有限公司 | Bill sharing information processing method and system |
| CN106067102A (en) * | 2016-05-24 | 2016-11-02 | 北京京东尚科信息技术有限公司 | The optimization method of layout for storekeeping and optimization device |
| CN107563702A (en) * | 2017-09-14 | 2018-01-09 | 北京京东尚科信息技术有限公司 | Commodity storage concocting method, device and storage medium |
| CN107784535A (en) * | 2016-08-24 | 2018-03-09 | 苏州乐聚堂电子科技有限公司 | Networked shopping system and method |
| CN110298615A (en) * | 2019-06-26 | 2019-10-01 | 网易无尾熊（杭州）科技有限公司 | For selecting method, apparatus, medium and the calculating equipment of the kinds of goods in warehouse |
| CN111292048A (en) * | 2020-03-31 | 2020-06-16 | 广东奥园奥买家电子商务有限公司 | Commodity ex-warehouse method and device for order splitting order of E-commerce platform |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140089133A1 (en) * | 2012-09-27 | 2014-03-27 | Wal-Mart Stores, Inc. | Techniques for determining substitutes for purchased items in a purchase history of a user |
| US20140156392A1 (en) * | 2012-12-05 | 2014-06-05 | Myworld, Inc. | Commerce System and Method of Controlling the Commerce System Using Budgets for Personalized Shopping Lists |
| US20150006285A1 (en) * | 2013-06-27 | 2015-01-01 | Telefonica Digital Espana, S.L.U. | Method and system for providing information regarding items in a retail store and computer programs thereof |
-
2020
- 2020-06-19 CN CN202010568808.5A patent/CN111738657B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679516A (en) * | 2012-09-25 | 2014-03-26 | 付艳超 | Price comparison system based on shopping cart and method thereof |
| CN105913301A (en) * | 2016-04-08 | 2016-08-31 | 珠海优特电力科技股份有限公司 | Bill sharing information processing method and system |
| CN106067102A (en) * | 2016-05-24 | 2016-11-02 | 北京京东尚科信息技术有限公司 | The optimization method of layout for storekeeping and optimization device |
| CN107784535A (en) * | 2016-08-24 | 2018-03-09 | 苏州乐聚堂电子科技有限公司 | Networked shopping system and method |
| CN107563702A (en) * | 2017-09-14 | 2018-01-09 | 北京京东尚科信息技术有限公司 | Commodity storage concocting method, device and storage medium |
| CN110298615A (en) * | 2019-06-26 | 2019-10-01 | 网易无尾熊（杭州）科技有限公司 | For selecting method, apparatus, medium and the calculating equipment of the kinds of goods in warehouse |
| CN111292048A (en) * | 2020-03-31 | 2020-06-16 | 广东奥园奥买家电子商务有限公司 | Commodity ex-warehouse method and device for order splitting order of E-commerce platform |
Non-Patent Citations (4)
| Title |
|---|
| Redesigning promotion strategy for e-commerce competitiveness through pricing and recommendation;Jiang YC等;《International Journal of Production Economics》;20150930;第167卷;全文 * |
| Utilize Item Correlation to Improve Aggregate Diversity for Recommender Systems;Liu yezheng等;《IEEE》;20170302;全文 * |
| 基于决策过程的个性化推荐系统设计;罗亚;《计算机工程与应用》;20170731(第14期);全文 * |
| 考虑商品重复购买周期的协同过滤推荐方法改进;张志清等;《武汉科技大学学报》;20170831(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111738657A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Supplier selection using consistent fuzzy preference relations | |
| Sadeghi et al. | An AHP decision making model for optimal allocation of energy subsidy among socio-economic subsectors in Iran | |
| Wang et al. | A strategy-oriented operation module for recommender systems in E-commerce | |
| CN113553540A (en) | Commodity sales prediction method | |
| CN111008858A (en) | Commodity sales prediction method and system | |
| CN102282551A (en) | Automated decision support for pricing entertainment tickets | |
| CN101785023A (en) | Article residual value predicting device | |
| Zhang et al. | Learning to select supplier portfolios for service supply chain | |
| CN108364191A (en) | Top-tier customer Optimum Identification Method and device based on random forest and logistic regression | |
| JP2005174313A (en) | Method and apparatus for calculating economic value of patent or technology | |
| CN115131101A (en) | Individualized intelligent recommendation system for insurance products | |
| CN111144986A (en) | Commodity recommendation method and device for social e-commerce website based on sharing behavior | |
| CN111192161A (en) | Electric power market trading object recommendation method and device | |
| Li et al. | Demand prediction, predictive shipping, and product allocation for large-scale e-commerce | |
| CN113988951A (en) | Commodity recommendation learning model construction method based on tensor decomposition and collaborative filtering | |
| CN115249164A (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
| CN113850654A (en) | Training method of item recommendation model, item screening method, device and equipment | |
| CN111177581A (en) | Multi-platform-based social e-commerce website commodity recommendation method and device | |
| CN115841345B (en) | Cross-border big data intelligent analysis method, system and storage medium | |
| CN111738657B (en) | Logistics bill splitting optimization method and system based on substitute recommendation | |
| CN112967108A (en) | BP-ANN-based electric power retail package searching and sequencing simulation system and method | |
| CN114418602A (en) | Online retailer product inventory decision-making method and system based on demand prediction | |
| Chen et al. | Managing the personalized order-holding problem in online retailing | |
| US20210027317A1 (en) | Inventory and structure finder | |
| CN114445099A (en) | Returned freight insurance pricing method, pricing system and construction method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |