Cross-border e-commerce order management method and system

Abstract

The invention provides a cross-border e-commerce order management method and system, and the method carries out the classification, scoring and sorting of different orders based on a fuzzy hierarchy method, and comprises the following steps: obtaining the information of a login user, and carrying out the login recognition of the login user through an encryption method; customer data of a sales end are collected, different order indexes are classified, and each index of each order is scored; performing priority ranking on the orders based on a fuzzy analytic hierarchy process according to different order index classification scoring results; and according to the order priority ranking sequence, sequentially delivering goods from the most emergency order to the least emergency order. According to the invention, rapid processing of orders and efficient distribution of logistics can be realized, and unified high-quality services can be provided globally, so that the satisfaction and loyalty of users are enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of e-commerce order management, and specifically relates to a cross-border e-commerce order management method and system. Background Art

[0002] With the rapid development of network technology, cross-border e-commerce has become a key link in global transactions. With the rise of cross-border e-commerce, enterprises have been able to have a larger market space and more abundant business opportunities, but this has also triggered many new challenges and competition. In this context, user shopping experience satisfaction has become a key factor in the operation of cross-border e-commerce. Through information technology and Internet platforms, order platforms and other platforms can achieve information sharing, collaboration and communication between different participants. In cross-border e-commerce, information collaboration can help enterprises and suppliers, logistics companies, payment institutions, customs and other parties to achieve real-time information sharing and collaborative decision-making, thereby improving operational efficiency, reducing costs and enhancing user experience. At the same time, the service conditions and indicators of upstream and downstream platforms such as logistics platforms and supply platforms that successfully trade orders during the shopping process will also affect the user experience.

[0003] The success of cross-border e-commerce is inseparable from a good service management system. With the continuous expansion of the cross-border e-commerce market, the service management system also needs to be continuously improved and developed in a coordinated manner. Cross-border e-commerce companies need to establish unified service standards and processes. Therefore, it is necessary to continue to establish a cross-border e-commerce order management method and system, which can couple and coordinate different evaluation indicators including the frequency of repeated orders for order purchases, the timeliness of logistics delivery, and the profit of purchased goods, and effectively sort the delivery order of multiple consumers' new orders within a certain time interval, and then effectively improve the user shopping experience satisfaction of the cross-border shopping platform by shipping according to the sorting. Summary of the invention

[0004] The present invention aims to solve the above-mentioned defects and provides a cross-border e-commerce order management method and system. The present invention can realize rapid order processing and efficient logistics distribution, so as to provide uniform high-quality services worldwide, thereby enhancing user satisfaction and loyalty.

[0005] The present invention provides the following technical solution: a cross-border e-commerce order management method, which classifies different orders based on a fuzzy hierarchy process and sorts them after scoring, comprising the following steps:

[0006] S1: Obtain the login user information and use encryption method to identify the login user;

[0007] S2: Collect customer data from the sales side, classify different order indicators, and score each indicator of each order;

[0008] S3: Based on the classification and scoring results of different order metrics, prioritize the orders using the fuzzy analytic hierarchy process.

[0009] S4: Ship the orders in sequence from the most urgent to the least urgent according to the order priority ranking.

[0010] As a preferred limitation of the present invention, the logged-in user in step S1 is identified using an encryption method, which includes the following steps:

[0011] S11: Obtain the user login information.

[0012] S12: Select a first prime number g and a second prime number d, where g ≠ d and g - 1 is divisible by d; then select an integer h and calculate L = h (g-1) / d mod g, where the integer h satisfies represents the real number field where the first prime number g is located; and L > 1, and use (g, d, h, L) as the public key.

[0013] S13: Select a random number v as the private key, 0 < v < d, and encrypt the login information Q to be signed. The encrypted digest Y of the real signal = L Q mod p;

[0014] Calculate the digital signature (D, F)

[0015] D = (h v mod g) mod d;

[0016] F = v -1 [H(Q) + D × Q] mod d;

[0017] where H(Q) is the one-way trapdoor function H(·) calculated for the login information Q to be signed.

[0018] where D is the first digital signature information and F is the second digital signature information.

[0019] S14: Send the calculated digital signature (D, F) as an authentication protocol to the mobile device used by the logged-in user, and send the public key (g, d, h, L) to the central control module.

[0020] The central control module decrypts the digital signature (D, F) received as an authentication protocol using the received public key (g, d, h, L).

[0021] As a preferred limitation of the present invention, step S15 includes:

[0022] S151: Calculate the first intermediate decryption value λ:

[0023] λ=F -1 mod d;

[0024] S152: Calculate the second intermediate decryption value σ according to the first intermediate decryption value λ:

[0025] σ=[λ×H(Q)]mod d;

[0026] S153: Calculate the third intermediate decryption value β:

[0027] β = (F × λ) mod d;

[0028] S154: Calculate the decrypted value z according to the calculation result of S152 and the calculation result of S153:

[0029] z=[(h σ Y β mod g)]mod d;

[0030] When the decrypted value z=D, the authentication digital signature (D, F) is valid, and the central control module allows the user to log in; otherwise, the authentication digital signature (D, F) is invalid, and the central control module does not allow the user to log in.

[0031] As a preferred limitation of the present invention, the different order indicators in step S2 include: order goods size C1, customer required time limit C2, order profit C3, and number of previous purchases by the purchase order customer C4;

[0032] The classification rules of different order indicators are as follows:

[0033]

[0034] Among them, C 11 For small items, C 12 For medium-sized products, C 13 For large items; i and V i are the weight and volume of the item in the i-th order respectively; W s , W q V are weight thresholds for small and medium-sized goods respectively; s 、V q are the volume thresholds for small and medium-sized goods respectively; BS(C 11 )=0.70~0.90;BS(C 12 )=0.40~0.60;BS(C 13 )=0.35~0.1;

[0035] BS(y) is the fuzzy scoring function for the independent variable y;

[0036]

[0037] Among them, C 21 For normal time limit orders, C 22 For expedited orders, C 23 For orders with overdue validity period; T i T is the time required for the i-th order customer to receive the goods; o , T U They are the delivery time of ordinary orders and the delivery time of expedited orders; BS (C 21 )=0.45~0.65;BS(C 22 )=0.85~1.0;BS(C 23 )=0.25~0.01;

[0038]

[0039] Among them, C 31 For high-profit orders, C 32 For medium profit orders, C 33 For low-profit orders; i is the total cost of the i-th order transaction, p b 、p a are the medium cost threshold for order execution and the high cost threshold for order execution respectively; BS(C 31 )=0.75~1.0;BS(C 32 )=0.55~0.70;BS(C 33 )=0.35~0.15;

[0040]

[0041] Among them, C 41 For high-frequency customers, C 42 For medium-frequency purchase customers, C 43 For low frequency purchasing customers; i is the frequency of the customer of the i-th order purchasing the same product up to now; f ∈ 、f δ are the high frequency threshold and medium frequency threshold of the target product being purchased; BS(C 41 )=0.75~0.95;BS(C 42 )=0.5~0.75;BS(C 43 )=0.15~0.05.

[0042] As a preferred limitation of the present invention, the S2 step comprises:

[0043] S21: Give three scores for each indicator; form the nth indicator score set R for the i-th order in : R in =(BS in,1 (C n ),BS in,2 (C n ),BS in,3 (C n ));BS in BS(y) is the fuzzy scoring function for the nth indicator of the i-th order; BS in,1 (C n ), BS in,2 (C n ) and BS in,3 (C n ) are the first scoring function value, the second scoring function value and the third scoring function value for the nth indicator of the ith order; i = 1, 2, ..., E; and the nth indicator scoring set R for the ith order in Matrix processing in =Matrix(R in )=[BS in,1 (C n )BS in,2 (C n )BS in,3 (C n )];

[0044] S22: Matrix the four indicator score sets of the i-th order to obtain the score matrix r of the i-th order i : Matrix is ​​a matrix processing function.

[0045] As a preferred limitation of the present invention, the order priorities are sorted based on the fuzzy analytic hierarchy process in step S3, including:

[0046] S31: Calculate the importance of the mth indicator relative to the nth indicator based on the naive Bayes method mn ;m=1,2,3,4; n=1,2,3,4; m≠n; A mn =(l mn ,k mn ,u mn ), l mn , k mn and u mn are the minimum possible value, the most likely value, and the maximum possible value of the mth indicator relative to the nth indicator, respectively;

[0047] S32: Construct fuzzy judgment matrix A:

[0048]

[0049] S33: Calculate all important triangular fuzzy numbers A of the mth row of the fuzzy matrix mn The geometric mean of is taken as the fuzzy weight S of the mth index. m :

[0050]

[0051] Among them, l m is the minimum possible weight coefficient of the mth indicator, k m is the most likely weight coefficient of the mth indicator, u mn is the maximum possible weight coefficient of the mth indicator;

[0052] S34: Fuzzy weight S for the mth indicator m Perform normalization processing to obtain the normalized fuzzy weight matrix of the mth indicator

[0053] Arrange the normalized fuzzy weight matrices of the m indicators horizontally from top to bottom to form the total fuzzy weight matrix

[0054]

[0055] S35: Using the fuzzy comprehensive evaluation method, the total fuzzy weight matrix of step S34 is adopted The rating matrix r of the i-th order obtained in step S22 i Perform matrix multiplication and take the diagonal of the result to obtain the order fuzzy score matrix of the i-th order diag is a function that takes diagonal elements to form a row matrix; where is the total fuzzy weight matrix The transposed matrix of lw i , kw i and uw i are the minimum possible score, the most likely score, and the maximum possible score of the i-th order respectively;

[0056] S36: Further calculate each element W in the order fuzzy score matrix W i The possible scores in are summed up and defuzzified to get the total score of the i-th order:

[0057]

[0058] S37: Sort the total scores of the E orders in descending order, and ship the orders in the order of sorting. The orders with higher scores are ranked higher and are shipped earlier.

[0059] As a preferred limitation of the present invention, in the step S31, the importance triangular fuzzy number A of the mth index relative to the nth index is calculated based on the naive Bayes method. mn The method is as follows:

[0060] S31: Obtain the customer history information collected by the sales end in step S2, and construct a consumer purchase feature vector X = {X1, X2, ..., X j ,…,X J}, j = 1, 2, ..., J; X j is the jth consumer feature vector;

[0061] S32: Calculate the mth index C by performing statistical analysis on the historical data of consumer purchase feature vectors m Relative to the nth index C n The conditional probability P(C m |C n ):

[0062]

[0063] Among them, P(C n |C m ) is the assumed indicator C m In important cases, indicator C n Also important is the probability, which can be obtained from historical data; P(C m ) is the indicator C m The prior probability of C is independent of other indicators. m The probability of being considered important; P(C n ) is the indicator C m The prior probability of C is that regardless of the importance of other indicators, n important probability;

[0064] S3: Based on the calculation result of step S32, calculate the minimum possible value l of the mth indicator relative to the nth indicator mn , the most likely value k mn and the maximum possible value u mn :

[0065] l mn =min(P(C m |C n ));k mn =mode(P(C m |C n ));u mn =max(P(C m |C n ));

[0066] min(P(C m |C n )) is used to calculate P(C m |C n ) The function with the smallest value among the results; mode(P(C m |C n )) is used to calculate P(C m |C n ) The function of the result that appears most often in the result; max(P(C m |C n )) is used to calculate P(C m |C n )The function with the largest value among the results.

[0067] The present invention also provides a cross-border e-commerce order management system as described above, the system includes a user login mobile terminal and a central control module, the system also includes: a login encryption identification module, an order classification and scoring module and an order priority sorting module;

[0068] The login encryption identification module is used to obtain the login user information and perform login identification on the login user using an encryption method;

[0069] The order classification and scoring module is used to collect customer data from the sales end, classify different order indicators, and score each indicator of each order;

[0070] The order priority sorting module is used to sort the orders according to the classification and scoring results of different order indicators based on the fuzzy hierarchical analysis method;

[0071] The central control module is used to receive the recognition result of the login encryption recognition module, and then control whether to allow the logged-in user; and is used to receive the sorting order result of the order priority sorting module, and is used to control the delivery of orders in order from the most urgent to the least urgent order according to the order priority sorting order.

[0072] The beneficial effects of the present invention are:

[0073] 1. The present invention uses big data analysis technology to mine and analyze massive amounts of cross-border e-commerce data, obtain information such as consumers' purchasing habits, preferences and needs, provide companies with accurate market positioning and product recommendations, and improve sales conversion rates and customer satisfaction. Through digital technology, visual management of the supply chain is achieved, including supplier selection and management, procurement and inventory optimization, logistics and distribution coordination, etc. Through digital supply chain management, costs can be reduced, efficiency can be improved, and better products and services can be provided.

[0074] 2. The method provided by the present invention can help sellers monitor and analyze data from all links of the successful order transaction chain in real time, including not only data from current links such as inventory levels, order status, logistics and transportation, and supplier performance, but also deep-dig historical data to find potential patterns. By aggregating, classifying, and deeply mining and analyzing these data, digital personalization technology can provide customers with a highly personalized online experience, thereby improving customer stickiness and loyalty.

[0075] 3. By analyzing the browsing and purchasing behaviors of customers on websites or apps, as well as their personal data based on content and interests, the present invention can allow companies to "label" each customer, conduct refined classification, and determine their characteristics, preferences, and potential needs. Then, companies can accurately push personalized goods or services that match the needs of different customers based on user portraits, and provide highly targeted product pages and activities. At the same time, customers can also experience the merchant's ultimate service and efficient order processing during the shopping process and after-sales service.

[0076] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] The present invention will be described in more detail below based on embodiments and with reference to the accompanying drawings, wherein:

[0078] Figure 1 A schematic diagram of the process flow of the cross-border e-commerce order management method provided by the present invention;

[0079] Figure 2 A schematic diagram of the principle of a method for user login encryption and decryption verification login in an embodiment of the present invention;

[0080] Figure 3 A bar chart showing the number of orders for different types of goods per year according to the present invention;

[0081] Figure 4 A bar chart showing the total order consumption amount per month for the year according to the embodiment of the present invention;

[0082] Figure 5 A schematic diagram of the structure of a cross-border e-commerce order management system provided by the present invention;

[0083] Figure 6 A statistical bar chart of order demand after three months of implementing different methods for order delivery sorting in the comparative example of the present invention; DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] like Figure 1 As shown, a cross-border e-commerce order management method provided by the present invention is provided. The method classifies different orders based on the fuzzy hierarchy process and sorts them after scoring, and includes the following steps:

[0086] S1: Obtain the login user information and use encryption method to identify the login user;

[0087] S2: Collect customer data from the sales side, classify different order indicators, and score each indicator of each order;

[0088] S3: Prioritize the orders based on the fuzzy analytic hierarchy process according to the classification and scoring results of different order indicators; the priority sorting of orders in this step refers to the process of analyzing all orders that consumers have paid but not shipped within a certain period of time according to historical data, assigning values ​​and classification scores to different orders, and further using the fuzzy analytic hierarchy process to prioritize them. The interval time can be changed according to the needs of decision makers, and can be 12h, 24h, 36h, 48h or 72h, etc.;

[0089] S4: Orders are shipped in order of priority from the most urgent to the least urgent.

[0090] Furthermore, in order to effectively audit the user login information and prevent unregistered users from logging into other registered accounts or hackers from cracking passwords to log in, it is necessary to encrypt and decrypt the login information entered by the user. Step S1 uses an encryption method to log in and identify the logged-in user, including the following steps:

[0091] S11: Figure 2 As shown in (a), the historical login user digitally signs his login information when logging in as a signer, and the central control module obtains the user login information as a signature verifier;

[0092] S12: Select a first prime number g and a second prime number d, g≠d, and g-1 is divisible by d; then select an integer h, and calculate L=h (g-1) / d mod g, the integer h satisfies Indicates the real number field where the first prime number g is located; and L>1, with (g, d, h, L) as the public key;

[0093] S13: Select a random number v as the private key, where 0 < v < d, and encrypt the login information Q to be signed. The encrypted digest Y of the real signal = L Q mod p;

[0094] Calculate the digital signature (D, F)

[0095] D = (h v mod g) mod d;

[0096] F = v -1 [H(Q) + D × Q] mod d;

[0097] where H(Q) is the one-way trapdoor function H(·) calculated for the login information Q to be signed;

[0098] where D is the first digital signature information and F is the second digital signature information;

[0099] S14: Send the calculated digital signature (D, F) as an authentication protocol to the mobile device used by the logged-in user, and send the public key (g, d, h, L) to the central control module;

[0100] S15: The mobile device used by the logged-in user sends the received digital signature (D, F) as an authentication protocol to the central control module. Further, as shown in Figure 2 (b), the central control module, as the verifier, decrypts the digital signature (D, F) as an authentication protocol using the received public key (g, d, h, L).

[0101] Step S15 includes:

[0102] S151: Calculate the first intermediate decryption value λ:

[0103] λ = F -1 mod d;

[0104] S152: Calculate the second intermediate decryption value σ based on the first intermediate decryption value λ:

[0105] σ = [λ × H(Q)] mod d;

[0106] S153: Calculate the third intermediate decryption value β:

[0107] β = (F × λ) mod d;

[0108] S154: Calculate the decryption value z based on the calculation results of S152 and S153:

[0109] z = [(h σ Y βmod g)]mod d;

[0110] like Figure 2 As shown in (b), when the decrypted value z=D, the authentication digital signature (D, F) is valid, and the central control module allows the user to log in; otherwise, the authentication digital signature (D, F) is invalid, and the central control module does not allow the user to log in.

[0111] like Figure 2 As shown in (b), when the decrypted value z=D, the central control module, as the verifier, authenticates that the digital signature (D, F) of the logged-in user DE is valid, and the identity authentication of the logged-in user is successful, allowing the user to enter the system for subsequent data cloud download, search, and historical order information statistics operations; otherwise, the central control module, as the verifier, authenticates that the digital signature (D, F) of the logged-in user is invalid, and the identity authentication of the user fails, and the user is not allowed to enter the system for data search and statistical operations.

[0112] The principle of verifying user login information by the asymmetric encryption method based on a one-way trapdoor function provided by the present invention is that the public key is published so that multiple login users do not need to decrypt one by one individually, and the login user information that needs to be authenticated is quickly and uniformly encrypted by the public key, and the private key is sent separately to the central control module for simultaneous and uniform decryption of the data information with digital signatures fed back by the user. This method not only ensures the confidentiality of the data, but also has the ability of fast digital signature, which can be used to quickly and uniformly verify the authenticity and integrity of multiple data. Using digital signatures as a security protocol plays a key role in preventing intruders from easily decrypting one-way. The protocol uses encryption technology to encrypt the login information Q, thereby effectively preventing man-in-the-middle attacks and eavesdropping. It provides a secure communication channel between the application layer and the transport layer to ensure that the data is not tampered with or leaked during transmission.

[0113] In order to count the sales data of sales-end customers under different evaluation indicators, it is necessary to clarify the indicators and assign scores to customer orders with different indicators to facilitate subsequent calculations of classified statistics. Therefore, as another preferred embodiment of the present invention, the different order indicators in step S2 include: order goods size C1, customer required time limit C2, order profit C3, and the number of previous purchases by the customer who purchased the order C4;

[0114] The classification rules for different order indicators are as follows:

[0115]

[0116] Among them, C 11 For small items, C 12 For medium-sized products, C 13 For large items; i and Vi are the weight and volume of the item in the i-th order respectively; W s , W q V are weight thresholds for small and medium-sized goods respectively; s 、V q are the volume thresholds for small and medium-sized goods respectively; BS(C 11 )=0.70~0.90;BS(C 12 )=0.40~0.60;BS(C 13 )=0.35~0.1;

[0117] BS(y) is the fuzzy scoring function for the independent variable y;

[0118]

[0119] Among them, C 21 For normal time limit orders, C 22 For expedited orders, C 23 For orders with overdue validity period; T i T is the time required for the i-th order customer to receive the goods; o , T U They are the delivery time of ordinary orders and the delivery time of expedited orders; BS (C 21 )=0.45~0.65;BS(C 22 )=0.85~1.0;BS(C 23 )=0.25~0.01; Generally, the normal time limit is 3-4 days to receive the goods, the expedited time limit is 1-2 days to receive the goods, and the overdue time limit is more than 5 days or the customer does not require the goods to be received within the normal time limit;

[0120]

[0121] Among them, C 31 For high-profit orders, C 32 For medium profit orders, C 33 For low profit orders; i is the total cost of the i-th order transaction, p b 、p a are the medium cost threshold for order execution and the high cost threshold for order execution respectively; BS(C 31 )=0.75~1.0;BS(C 32 )=0.55~0.70;BS(C 33 )=0.35~0.15;

[0122]

[0123] Among them, C 41 For high-frequency customers, C42 For medium-frequency purchase customers, C 43 For low frequency purchasing customers; i is the frequency of the customer of the i-th order purchasing the same product up to now; f ∈ 、f δ are the high frequency threshold and medium frequency threshold of the target product being purchased; BS(C 41 )=0.75~0.95;BS(C 42 )=0.5~0.75;BS(C 43 )=0.15~0.05.

[0124] like Figure 3 As shown in the figure, the customer order data obtained from the sales end in 2022, according to different commodity types and purchasing ages, different colored columns represent different purchasing ages, the left side is the commodity type segmentation, and the horizontal axis is the percentage of the purchase quantity of this age group in one year to the total purchase quantity of this commodity by different age groups (the purchase frequency can then be calculated). Figure 4 As shown, it is the statistics of the purchase order amount of different age groups when conducting transactions through this platform every month in 2022. The vertical axis is the different order amount ranges, and the horizontal axis is the percentage of purchase transactions of different age groups to the total amount of purchases of this kind of goods by different age groups.

[0125] Step S2 includes the following steps to assign scores to sales orders under various indicators:

[0126] S21: Give three scores for each indicator; form the nth indicator score set R for the i-th order in : R in =(BS in,1 (C n ),BS in,2 (C n ),BS in,3 (C n ));BS in BS(y) is the fuzzy scoring function for the nth indicator of the i-th order; BS in,1 (C n ), BS in,2 (C n ) and BS in,3 (C n ) are the first scoring function value, the second scoring function value and the third scoring function value for the nth indicator of the ith order; i = 1, 2, ..., E; and the nth indicator scoring set R for the ith order in Matrix processing in =Matrix(R in )=[BS in,1(C n )BS in,2 (C n )BS in,3 (C n )];

[0127] S22: Matrix the four indicator score sets of the i-th order to obtain the score matrix r of the i-th order i : Matrix is ​​a matrix processing function.

[0128] More preferably, in order to sort the delivery priority of the order placed by the customer this time based on the valued scoring marks of the historical orders of different customers, it is necessary to further sort the delivery order of all orders in the past 24 hours based on the valued scoring marks. Therefore, as another preferred embodiment of the present invention, in step S3, the orders are prioritized based on the fuzzy hierarchical analysis method, including:

[0129] S31: Calculate the importance of the mth indicator relative to the nth indicator based on the naive Bayes method mn ;m=1,2,3,4; n=1,2,3,4; m≠n; A mn =(l mn ,k mn ,u mn ), l mn , k mn and u mn are the minimum possible value, the most likely value, and the maximum possible value of the mth indicator relative to the nth indicator, respectively;

[0130] S32: Construct fuzzy judgment matrix A:

[0131]

[0132] S33: Calculate all important triangular fuzzy numbers A of the mth row of the fuzzy matrix mn The geometric mean of m :

[0133]

[0134] Among them, l m is the minimum possible weight coefficient of the mth indicator, k m is the most likely weight coefficient of the mth indicator, u mn is the maximum possible weight coefficient of the mth indicator;

[0135] S34: Fuzzy weight S for the mth indicator m Perform normalization processing to obtain the normalized fuzzy weight matrix of the mth indicator

[0136] Arrange the normalized fuzzy weight matrices of the m indicators horizontally from top to bottom to form the total fuzzy weight matrix

[0137] S35, using the fuzzy comprehensive evaluation method, adopting the total fuzzy weight matrix of step S34 The rating matrix r of the i-th order obtained in step S22 i Perform matrix multiplication and take the diagonal of the result to obtain the order fuzzy score matrix of the i-th order diag is a function that takes diagonal elements to form a row matrix; where is the total fuzzy weight matrix The transposed matrix of i , kw i and uw i are the minimum possible score, the most likely score, and the maximum possible score of the i-th order respectively;

[0138]

[0139] The above lw i , kw i and uw i In the calculation formula, m = n, such as when n = 1, m = 1; when n = 2, m = 2; and In lw i , kw i and uw i The calculation formula is a fixed value determined by the calculation result and is not affected by the change of n from 1 to 4.

[0140] S36, further calculate each element W in the order fuzzy score matrix W i The possible scores in are summed up and defuzzified to get the total score of the i-th order:

[0141]

[0142] S37. Sort the total scores of the E orders in descending order, and ship them in the order of sorting. The higher the score, the higher the ranking, and the earlier the order is shipped.

[0143] The order priority sorting based on fuzzy analytic hierarchy process combines the advantages of fuzzy mathematics and analytic hierarchy process to solve complex decision-making problems caused by uncertainty and subjective judgment. The method provided by the present invention uses a naive Bayesian method to calculate three triangular fuzzy numbers of the importance of two different indicators for shipment sorting. mn , k mn and u mn Based on this, the fuzzy weight is further calculated and normalized to obtain the total fuzzy weight matrix of a certain indicator. Then the rating matrix r of the i-th order i Perform matrix multiplication and take the matrix elements after the diagonal of the result for defuzzification to obtain the total score of the i-th order. i , which can be used to solve the uncertainty in decision makers' evaluation of order importance. For example, when comparing the priorities of two orders, decision makers can use triangular fuzzy numbers of different indicators to express uncertain preferences; further, it is necessary to compare which indicator is more important for different orders, so when facing two indicators, which indicator should be given priority in order delivery; for example, the cost of the i-th order is also high, but customers buy frequently (high purchase frequency), and the cost of the j-th order is relatively low, but customers do not buy frequently (low purchase frequency), then it is necessary to ship the i-th order first or the j-th order first, and it is necessary to set the indicator preference and then sort the order delivery priority accordingly. Using fuzzy comprehensive evaluation, the fuzzy score of each order under different criteria is calculated, and by removing the model, it is particularly suitable for the problem of order priority sorting in the process of cross-border e-commerce order management, helping decision makers sort the order priorities under multiple criteria.

[0144] Furthermore, in step S31, the importance triangular fuzzy number A of the mth indicator relative to the nth indicator is calculated based on the naive Bayes method. mn The method is as follows:

[0145] S31: Obtain the customer history information collected by the sales end in step S2 and construct the consumer purchase feature vector X = {X1, X2, ..., X j ,…,X J}, j = 1, 2, ..., J; X j is the jth consumer feature vector;

[0146] For example: X = {X1, X2, X3, X4, X5}, X1 is the time of the consumer's most recent purchase (e.g. "last month", "last three months", etc.), X2 is the consumer's purchase frequency (e.g. high frequency, medium quality, low frequency, etc.), X3 is the price range of the product, in yuan; X4 is whether the consumer has purchased the product in the past, and X5 is the category of the product (e.g. electronic products, daily necessities, etc.);

[0147] S32: Calculate the mth index C by performing statistical analysis on the historical data of consumer purchase feature vectors m Relative to the nth index C n The conditional probability P(C m |C n ):

[0148]

[0149] Among them, P(C n |C m ) is the assumed indicator C m In important cases, indicator C n It is also important to consider the possibility that this can be obtained from historical data; P(C n |C m )=P(C n1 |C m )+P(C n2 |C m )+P(C n3 |C m ); P(C m ) is the indicator C m The prior probability of C is independent of other indicators. m The probability of being considered important, P(C m )=P(C m1 )+P(C m2 )+P(C m3 ); P(C n ) is the indicator C m The prior probability of C is that no matter how important other indicators are, n The important probability, P(C m )=P(C n1 )+P(C n2 )+P(C n3 );

[0150] S33: Based on the calculation result of step S32, calculate the minimum possible value l of the mth indicator relative to the nth indicator mn , the most likely value k mn and the maximum possible value u mn :

[0151] l mn =min(P(C m |C n ));k mn =mode(P(C m |C n ));u mn =max(P(C m |Cn ));

[0152] min(P(C m |C n )) is used to calculate P(C m |C n ) is the function with the smallest value in the result, which is an estimate of the conditional probability under the most unfavorable circumstances and is used to represent the index C m Relative to index C n The lower bound of the relative importance of mode(P(C m |C n )) is used to calculate P(C m |C n ) is the function of the result that appears the most times in the result. This is an estimate of the most likely conditional probability, the most likely value calculated through historical data; max(P(C m |C n )) is used to calculate P(C m |C n ) is the function with the largest value in the result, which is an estimate of the conditional probability under the most optimistic situation and is used to represent the index C m Relative to index C n An upper bound on the relative importance of .

[0153] The present invention also provides a cross-border e-commerce order management system 1 using the above method, such as Figure 5 As shown, the system includes a user login mobile terminal and a central control module 2, and the system also includes: a login encryption identification module 3, an order classification and scoring module 41 and an order priority sorting module 42;

[0154] The login encryption identification module 3 is used to obtain the login user information and perform login identification on the login user using an encryption method;

[0155] The order classification and scoring module 41 is used to collect customer data from the sales end, classify different order indicators, and score each indicator of each order;

[0156] An order priority sorting module 42 is used to sort the orders according to the classification and scoring results of different order indicators based on the fuzzy hierarchical analysis method;

[0157] The central control module 2 is used to receive the recognition result of the login encryption recognition module, and then control whether the logged-in user is allowed; and is used to receive the sorting order result of the order priority sorting module, and is used to control the delivery of orders in order from the most urgent to the least urgent order according to the order priority sorting order.

[0158] The central control module of the system provided by the present invention is a processor, which has a memory inside, and the memory has a computer-readable medium, a communication interface 11 and a bus, and may also include a computer-readable program 21 stored in the memory and executable on the central control module 2 as a processor. The communication interface 11 may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the system and other electronic devices, and to realize the connection and communication between the internal components of the electronic devices.

[0159] The storage includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. The memory may be an internal storage unit of the system in some embodiments, such as a mobile hard disk of the system. The memory may also be an external storage device of the system in other embodiments, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the system. Further, the memory may also include both an internal storage unit of the system and an external storage device. The memory may be used not only to store application software and various types of data installed in the system, such as the code of program 12, etc., but also to temporarily store data that has been output or is to be output. The computer-readable program 21 stores at least one instruction, and the at least one instruction is executed by the central control module in the system to implement the cross-border e-commerce order management method including the above steps S1-S4.

[0160] In order to further illustrate with specific data that the method provided by the present invention uses the naive Bayes algorithm to calculate three fuzzy numbers for different order indicators, and then obtain the values ​​of each element in the fuzzy judgment matrix A of the importance of different order indicators, and then further calculate the total fuzzy weight matrix based on the fuzzy hierarchical analysis method The present invention provides the following embodiments for further calculating the detailed calculation steps of the final score of each order.

[0161] Example 1

[0162] S1: Obtain the login user information and use encryption method to identify the login user; the specific implementation scheme of this step is the same as the above steps S11-S15;

[0163] Step S2: The platform receives three orders in the past 24 hours, namely the first order O1, the second order O2 and the third order O3. The three orders need to be prioritized according to the following four indicators: the size of the goods C1, the time from placing the order to receiving the goods C2, the order profit C3 and the frequency of purchase of the goods by the customers placing the corresponding orders C4.

[0164] Step S21: Collect customer data from the sales end, classify different order indicators, and score each indicator of each of the three orders in turn:

[0165] For the three scoring results of order O1 with order size C1, BS 11,1 (C1), BS 11,2 (C1) and BS 11,3 (C1) are 0.75, 0.70 and 0.78 respectively. The three scoring results of the customer's requirement timeliness C2 are BS 12,1 (C2), BS 12,2 (C2) and BS 12,3 (C2) are 0.55, 0.47 and 0.60 respectively, the three scoring results of order profit C3 are 0.76, 0.83 and 0.91 respectively, and the three scoring results of the number of previous purchases by the customer who purchased the order C4 are 0.53, 0.61 and 0.72 respectively.

[0166] For order O2, the three scoring results of the order cargo size C1 were 0.33, 0.30 and 0.26 respectively, the three scoring results of the customer required time limit C2 were 0.87, 0.91 and 0.93 respectively, the three scoring results of the order profit C3 were 0.66, 0.58 and 0.70 respectively, and the three scoring results of the number of previous purchases by the purchasing order customer C4 were 0.365, 0.21 and 0.17 respectively.

[0167] For order O3, the three scoring results of the order cargo size C1 were 0.425, 0.478 and 0.519 respectively, the three scoring results of the customer required time limit C2 were 0.25, 0.18 and 0.13 respectively, the three scoring results of the order profit C3 were 0.37, 0.28 and 0.165 respectively, and the three scoring results of the number of previous purchases by the purchasing order customer C4 were 0.167, 0.23 and 0.125 respectively.

[0168] Therefore, order O1 is a small item that requires normal delivery time, the item purchased by the customer is a high-profit order after the transaction is completed, and the customer has previously purchased this item at a medium frequency;

[0169] Order O2 is a large-sized product that requires expedited delivery. The product purchased by the customer is a medium-profit order after the transaction is completed, and the customer has purchased this product frequently before.

[0170] Order O3 is a medium-sized product, the customer does not require a delivery time limit (that is, the product can be received beyond the delivery time limit), the product purchased by the customer is a low-profit order after the transaction is completed, and the customer has previously purchased this product at a low frequency;

[0171] The fuzziness score set of each order is shown in Table 1.

[0172] Table 1

[0173]

[0174] Step S22: The score set R of the first indicator C1 of the first order O1 11 =(0.75, 0.70, 0.78), the score set R of the second indicator C2 of the first order O1 12 =(0.55, 0.47, 0.60), the score set R of the third indicator C3 of the first order O1 13 =(0.76, 0.83, 0.91), the score set R of the fourth indicator C3 of the first order O1 14 =(0.53,0.61,0.72).

[0175]

[0176] The scoring sets of various indicators of the second order O2 and the third order O3 and the corresponding scoring matrices r2 and r3 can also be obtained from Table 1.

[0177] Step S3: Calculate the scores of the three orders based on different criteria based on the naive Bayes method, and construct the fuzzy judgment matrix A as follows:

[0178]

[0179] The first row in the judgment matrix A represents the relative importance of C1 and other indicators, the second row represents the relative importance of the time from placing an order to receiving the goods C2 and other indicators, the third row represents the importance of order profit C3 and other indicators, and the fourth row represents the frequency of purchase of the product by the corresponding ordering customers C4 and other indicators.

[0180] A 11 =(l 11 ,k 11 ,u 11 )=(1,1,1),A 12 =(l 12 ,k 12 ,u 12 )=(0.3,0.5,0.7), A 13=(l 13 ,k 13 ,u 13 )=(0.6,0.7,0.8),A 14 =(l 14 ,k 14 ,u 14 )=(0.4,0.6,0.9);

[0181] A 21 =(l 21 ,k 21 ,u 21 )=(1.4,2.0,3.0),A 22 =(l 22 ,k 22 ,u 22 )=(1,1,1),A 23 =(l 23 ,k 23 ,u 23 )=(0.7,0.8,0.9),A 24 =(l 24 ,k 24 ,u 24 )=(0.5,0.6,0.7);

[0182] A 31 =(l 31 ,k 31 ,u 31 )=(1.2,1.4,1.7),A 32 =(l 32 ,k 32 ,u 32 )=(1.1,1.25,1.4),A 33 =(l 33 ,k 33 ,u 33 )=(1,1,1),A 34 =(l 34 ,k 34 ,u 34 )=(0.3,0.4,0.5);

[0183] A 41 =(l 41 ,k 41 ,u 41 )=(0.9,1.0,1.1),A 42 =(l 42 ,k 42 ,u 42 )=(1.3,1.5,1.7),A 43 =(l 43 ,k43 ,u 43 )=(2.0,2.3,2.6),A 44 =(l 44 ,k 44 ,u 44 )=(1,1,1).

[0184] Step 4) Use the geometric mean method to calculate the fuzzy weights S1, S2, S3 and S4 of each indicator:

[0185] S1 is the fuzzy weight of indicator C1, S2 is the fuzzy weight of indicator C2, S3 is the fuzzy weight of C3, and S4 is the fuzzy weight of indicator C4;

[0186] S1=(l1,k1,u1)

[0187] =((1×0.3×0.6×0.4) 1 / 4 ,(1×0.5×0.7×0.6) 1 / 4 ,(1×0.7×0.8×0.9) 1 / 4 )

[0188] =(0.518,0.676,0.843)

[0189] l1=l 11 × 12 × 13 × 14 =(1×0.3×0.6×0.4) 1 / 4 =0.518

[0190] k1=k 11 ×k 12 ×k 13 ×k 14 =(1×0.5×0.7×0.6) 1 / 4 =0.676

[0191] u1=u 11 ×u 12 ×u 13 ×u 14 =(1×0.7×0.8×0.9) 1 / 4 =0.897

[0192] S2=(l2,k2,u2)

[0193] =((1.4×1×0.7×0.5) 1 / 4 ,(2.0×1×0.8×0.6) 1 / 4 ,(3.0×1×0.9×0.7) 1 / 4 )

[0194] =(0.837,0.990,1.173)

[0195] l2=l 21 ×l 22 ×l 23 ×l 24 =(1.4×1×0.7×0.5) 1 / 4 =0.837

[0196] k2=k 21 ×k 22 ×k 23 ×k 24 =(2.0×1×0.8×0.6) 1 / 4 =0.990

[0197] u2=U 21 ×u 22 ×u 23 ×u 24 =(3.0×1×0.9×0.7) 1 / 4 =1.173

[0198] S3=(l3,k3,u3)

[0199] =((1.2×1.1×1×0.3) 1 / 4 ,(1.4×1.25×1×0.4) 1 / 4 ,(1.7×1.4×1

[0200] ×0.5) 1 / 4 )=(0.793,0.915,1.044)

[0201] l3=l 31 ×l 32 ×l 33 ×l 34 =(1.2×1.1×1×0.3) 1 / 4 =0.793

[0202] k3=k 31 ×k 32 ×k 33 ×k 34 =(1.4×1.25×1×0.4) 1 / 4 =0.915

[0203] u3=u 31 ×u 32 ×u 33 ×u 34 =(1.7×1.4×1×0.5) 1 / 4 =1.044

[0204] S4=(l4,k4,u4)

[0205] =((0.9×1.3×2.0×1) 1 / 4 ,(1.0×1.5×2.3×1) 1 / 4 ,(1.1×1.7×2.6×1) 1 / 4 )

[0206] =(1.237,1.363,1.485)

[0207] l4=l 41 × 42 × 43 × 44 =(0.9×1.3×2.0×1) 1 / 4 =1.237

[0208] k4=k 41 ×k 42 ×k 43 ×k 44 =(1.0×1.5×2.3×1) 1 / 4 =1.363

[0209] u4=U 41 ×u 42 ×u 43 ×u 44 =(1.1×1.7×2.6×1) 1 / 4 =1.485

[0210] Step S33: Normalize the fuzzy weights S1, S2, S3 and S4 of each indicator. First, calculate the sum of the maximum possible weight coefficients of each indicator fuzzy weight. The sum of the most likely weight coefficients of each indicator's fuzzy weight

[0211] Step S34: The sum of the minimum possible weight coefficients of each indicator fuzzy weight Further calculate the normalized fuzzy weight matrix of the first indicator C1 The normalized fuzzy weight matrix of the second index C2 The normalized fuzzy weight matrix of the third index C3 is And the normalized fuzzy weight matrix of the second index C2

[0212]

[0213] ; The normalized fuzzy weight matrices of the four indicators are arranged horizontally from top to bottom to form a total fuzzy weight matrix

[0214] Its transposed matrix

[0215] Step S35 and step S36: Use the fuzzy comprehensive evaluation method, using the transposed matrix Perform matrix multiplication with the score matrices r1, r2 and r3 of the first order O1 to the third order O3 respectively and take the diagonal elements to form the matrix W1, and obtain the order fuzzy score matrix W1 of the first order = [0.463 0.641 1.011], that is, W1 is The diagonal elements of the calculation results form a 1×3 matrix;

[0216] Then lw1=0.463, kw1=0.641, uw1=1.011, defuzzify it, and get the total score of the first order Sorce1=(0.463+4×0.641+1.011)÷6=0.673;

[0217] For the second order O2, the transposed matrix is ​​also used Perform matrix multiplication with its rating matrix r2 and take the diagonal elements to form the matrix W2, and obtain the order fuzzy score matrix W2 = [0.412 0.487 0.682] of the second order O2; that is, W2 is The diagonal elements of the calculation results form a 1×3 matrix;

[0218] Then lw2=0.412,kw2=0.487,uw2=0.682,defuzzify it and get the total score of the first order Sorce2=(0.412+4×0.487+0.682)÷6=0.507;

[0219] For the third order O3, the transposed matrix is ​​also used Perform matrix multiplication with its score matrix r3 and take the diagonal elements to form the matrix W2, and obtain the order fuzzy score matrix W3 = [0.204 0.271 0.288] of the third order O3; that is, W2 is The diagonal elements of the calculation results form a 1×3 matrix;

[0220] Then lw3=0.204, kw3=0.271, uw3=0.288, defuzzify it, and get the total score of the first order Sorce3=(0.204+4×0.271+0.288)÷6=0.262;

[0221] Step S37: Since Sorce1>Sorce2>Sorce3, the orders are shipped in the order of the first order first, the second order secondly, and the third order lastly.

[0222] In order to clarify the customer experience after order sorting by the method provided by the present invention, the present invention provides the following comparative examples.

[0223] Comparative Example 1

[0224] After the order is sorted by the method provided by the present invention and a certain step is reduced, the goods are shipped. After executing various methods for 3 months, they are continued to be executed for 7 days. The customer purchase order volume within these 7 days is monitored to reflect the change in customer purchasing intention after executing various methods for a certain period of time. The height of each column is the order volume of a single day of the detection. For example, when the horizontal axis is 5, the corresponding column height is the order volume on the 5th day. Figure 6 The orange column in (a) shows the total order volume of various commodities purchased within 7 days after the order sorting method provided by the present invention is used. Figure 6 The blue column in (a) indicates that when the method provided by the present invention is used to sort the order of order delivery, all the important triangular fuzzy numbers A of the mth row of the fuzzy matrix are calculated in step S33. mn When , the geometric mean is not used, and the average of the triangular fuzzy numbers in the mth row is directly calculated, and the normalized fuzzy weight processing is not used;

[0225] That is, in step S33, the fuzzy weight S of the mth indicator is calculated. m =(l m ,k m ,u m )hour, The fuzzy weight matrix of the mth indicator obtained in step S34 as follows: The remaining steps are the same as those of the method provided by the present invention;

[0226] Figure 6 The green column in (a) indicates that Figure 6 The blue column in (a) represents the order management method. Steps S33-S34 are changed in the same way. Step S35 is the same, but step S36 is not defuzzified. lw is directly calculated. i , kw i and uw i The sum is the total score of the i-th order i , and sort multiple orders from large to small according to the total scores, and ship the orders according to the sorting.

[0227] Figure 6The orange column of (b) is relative to the order management method provided by the present invention (i.e. Figure 6 Based on the method represented by the yellow column in (a), in step S31, the importance triangular fuzzy number A of the mth index relative to the nth index is not calculated based on the prime Bayesian method. mn ; Figure 6 The orange column in (b) represents the method that calculates A in step S31. mn When A mn =(l mn ,k mn ,u mn ), Figure 6 (b) The k of the method mn For multiple orders with C m The number of times the average of the three scores exceeds 0.75 when the indicator is used is the same as the number of multiple orders with C n The ratio of the number of indicators with the average of three scores exceeding 0.75; mn For multiple orders with C m The minimum number of the three average scores for the indicator is the same as the number of C for multiple orders. n The ratio of the number of the smallest average of three scores when it is an indicator; u mn For multiple orders with C m When the indicator is used, the average of three scores is used for multiple orders. m The maximum number of the three averages is scored as an indicator and the C is used for multiple orders. n The ratio of the largest number of averages of three scores when it is an indicator;

[0228] Figure 6 The blue column method in (b) is based on Figure 6 Based on the method of the orange column in (b), in step S33, all important triangular fuzzy numbers A of the mth row of the fuzzy matrix are calculated. mn When , the geometric mean is not used, and the average of the triangular fuzzy numbers in the mth row is directly calculated, and the normalized fuzzy weight processing is not used;

[0229] That is, in step S33, the fuzzy weight S of the mth indicator is calculated. m =(l m ,k m ,u m )hour, The fuzzy weight matrix of the mth indicator obtained in step S34 as follows: The remaining steps are Figure 6 The method is the same for the orange column in (b);

[0230] Figure 6 The green column method in (b) is based on Figure 6Based on the green column in (b), the further step S35 is the same, but the step S36 does not perform defuzzification; lw is directly calculated i , kw i and uw i The sum is the total score of the i-th order i , and sort multiple orders from large to small according to the total scores, and ship the orders according to the sorting.

[0231] like Figure 6 (a) with Figure 6 (b) orange column comparison, using Figure 6 (a) After 3 months of using the method provided by the present invention, consumers' willingness to place orders is relative to Figure 6 The order delivery sequence of the orange column in (b) shows a stronger willingness to place orders and purchase goods on the platform under the order management controlled by this system. Figure 6 Although the blue column in (a) has a decrease in order volume compared to the yellow column in this figure, it is Figure 6 The order volume of the blue column in (b) still has an advantage. Figure 6 (a) and Figure 6 The same result is obtained when comparing the green columns in (b).

[0232] It can be seen that the importance of the mth index relative to the nth index is calculated based on the naive Bayesian method. mn It can more accurately determine the evaluation indicators of each order, and then effectively clarify which order indicator is more important in the delivery sorting process of each order, which is conducive to the accurate calculation of the final total score.

[0233] Figure 6 (a) were compared separately and found that based on Figure 6 (a) Orange column The order management method of the present invention in all steps S31-S37 manages the order after delivery for 3 months, compared with Figure 6 (a) The S33-S34 step of the blue column does not perform geometric mean calculation or normalization, which indicates a stronger purchase intention. Figure 6 (a) The difference between the green columns in steps S33-S34 and S36 indicates a stronger purchase intention. Figure 6 When the orange column, blue column, and green column in (b) are compared within the graph, Figure 6 (a) The results of the comparison within the figure are the same. Therefore, it is shown that the steps S33-S34 and S36 of the method of the present invention are important for the accuracy of the order delivery sequence sorting of the fuzzy hierarchy process.

[0234] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A cross-border e-commerce order management method, which classifies and scores different orders based on a fuzzy hierarchy process and sorts them, and is characterized by: The following steps are involved: S1: Obtain the login user information and use encryption method to identify the login user; S2: Collect customer data from the sales side, classify different order indicators, and score each indicator of each order; S3: According to the classification and scoring results of different order indicators, the orders are prioritized based on the fuzzy analytic hierarchy process; S4: Orders are shipped in order of priority from the most urgent to the least urgent.

2. The cross-border e-commerce order management method according to claim 1, characterized in that: The S1 step of logging in the user adopts an encryption method to perform login identification, including the following steps: S11: Obtain user login information; S12: Select a first prime number g and a second prime number d, g≠d, and g-1 is divisible by d; then select an integer h, and calculate L=h (g-1) / d mod g, the integer h satisfies Indicates the real number field where the first prime number g is located; and L>1, with (g, d, h, L) as the public key; S13: Select a random number v as the private key, where 0 < v < d, and encrypt the login information Q to be signed. The encrypted digest Y of the real signal = L Q mod p; Calculate digital signature (D, F) D=(h v mod g)mod d; F=v -1 [H(Q)+D×Q]mod d; Wherein, H(Q) is the one-way trapdoor function H(·) calculated for the login information Q to be signed; Wherein, D is the first digital signature information, and F is the second digital signature information; S14: Send the calculated digital signature (D, F) as an authentication protocol to the mobile terminal used by the logged-in user, and send the public key (g, d, h, L) to the central control module; S15: The digital signature (D, F) received by the mobile terminal used by the logged-in user as the authentication protocol is sent to the central control module, and the central control module uses the received public key (g, d, h, L) to decrypt the digital signature (D, F) as the authentication protocol.

3. The cross-border e-commerce order management method according to claim 2, characterized in that: The step S15 comprises: S151: Calculate the first intermediate decryption value λ: λ=F -1 mod d; S152: Calculate the second intermediate decryption value σ according to the first intermediate decryption value λ: σ=[λ×H(Q)]mod d; S153: Calculate the third intermediate decryption value β: β = (F × λ) mod d; S154: Calculate the decrypted value z according to the calculation result of S152 and the calculation result of S153: z=[(h σ Y β mod g)]mod d; When the decrypted value z=D, the authentication digital signature (D, F) is valid, and the central control module allows the user to log in; otherwise, the authentication digital signature (D, F) is invalid, and the central control module does not allow the user to log in.

4. The cross-border e-commerce order management method according to claim 1, characterized in that: The different order indicators in step S2 include: order goods size C1, customer required time limit C2, order profit C3, and the number of previous purchases by the order customer C4; The classification rules of different order indicators are as follows: Among them, C 11 For small items, C 12 For medium-sized products, C 13 For large items; i and V i are the weight and volume of the item in the i-th order respectively; W s , W q V are weight thresholds for small and medium-sized goods respectively; s 、V q are the volume thresholds for small and medium-sized goods respectively; BS(C 11 )=0.70~0.90;BS(C 12 )=0.40~0.60;BS(C 13 )=0.35~0.1; BS(y) is the fuzzy scoring function for the independent variable y; Among them, C 21 For normal time limit orders, C 22 For expedited orders, C 23 For orders with overdue validity period; T i T is the time required for the i-th order customer to receive the goods; o , T U They are the delivery time of ordinary orders and the delivery time of expedited orders; BS (C 21 )=0.45~0.65;BS(C 22 )=0.85~1.0;BS(C 23 )=0.25~0.01; Among them, C 31 For high-profit orders, C 32 For medium profit orders, C 33 For low profit orders; i is the total cost of the i-th order transaction, p b 、p a are the medium cost threshold for order execution and the high cost threshold for order execution respectively; BS(C 31 )=0.75~1.0;BS(C 32 )=0.55~0.70;BS(C 33 )=0.35~0.15; Among them, C 41 For high-frequency customers, C 42 For medium-frequency purchase customers, C 43 For low frequency purchasing customers; i is the frequency of the customer of the i-th order purchasing the same product up to now; f ∈ 、f δ are the high frequency threshold and medium frequency threshold of the target product being purchased; BS(C 41 )=0.75~0.95;BS(C 42 )=0.5~0.75;BS(C 43 )=0.15~0.

05.

5. The cross-border e-commerce order management method according to claim 4, characterized in that: The S2 step includes: S21: Give three scores for each indicator; form the nth indicator score set R for the i-th order in : R in =(BS in，1 (C n ), B.S. in，2 (C n ), B.S. in，3 (C n ));BS in BS(y) is the fuzzy scoring function for the nth indicator of the i-th order; BS in，1 (C n ), BS in，2 (C n ) and BS in，3 (C n ) are the first scoring function value, the second scoring function value and the third scoring function value for the nth indicator of the ith order respectively; i = 1, 2, ..., E; and the nth indicator scoring set R for the ith order in Matrix processing in =Matrix(R in )=[BS in，1 (C n ) BS in，2 (C n ) BS in，3 (C n )]; S22: Matrix the four indicator score sets of the i-th order to obtain the score matrix r of the i-th order i : Matrix is ​​a matrix processing function.

6. The cross-border e-commerce order management method according to claim 5, characterized in that: In the step S3, the orders are prioritized based on the fuzzy analytic hierarchy process, including: S31: Calculate the importance of the mth indicator relative to the nth indicator based on the naive Bayes method mn ;m=1,2,3,4;n=1,2,3,4;m≠n;A mn =(l mn , k mn ,u mn ), l mn , k mn and u mn are the minimum possible value, the most likely value, and the maximum possible value of the mth indicator relative to the nth indicator, respectively; S32: Construct fuzzy judgment matrix A: S33: Calculate all important triangular fuzzy numbers A of the mth row of the fuzzy matrix mn The geometric mean of is taken as the fuzzy weight s of the mth indicator. m : Among them, l m is the minimum possible weight coefficient of the mth indicator, k m is the most likely weight coefficient of the mth indicator, u mn is the maximum possible weight coefficient of the mth indicator; S34: Fuzzy weight S for the mth indicator m Perform normalization processing to obtain the normalized fuzzy weight matrix of the mth indicator Arrange the normalized fuzzy weight matrices of the m indicators horizontally from top to bottom to form the total fuzzy weight matrix S35: Using the fuzzy comprehensive evaluation method, the total fuzzy weight matrix of step S34 is adopted The rating matrix r of the i-th order obtained in step S22 i Perform matrix multiplication and take the diagonal of the result to obtain the order fuzzy score matrix of the i-th order diag is a function that takes diagonal elements to form a row matrix; where is the total fuzzy weight matrix The transposed matrix of lw i , kw i and uw i are the minimum possible score, the most likely score, and the maximum possible score of the i-th order respectively; S36: Further calculate each element W in the order fuzzy score matrix W i The possible scores in are summed up and defuzzified to get the total score of the i-th order: S37: Sort the total scores of the E orders in descending order, and ship the orders in the order of sorting. The orders with higher scores are ranked higher and are shipped earlier.

7. The cross-border e-commerce order management method according to claim 6, characterized in that: In the step S31, the importance triangular fuzzy number A of the mth index relative to the nth index is calculated based on the naive Bayes method. mn The method is as follows: S31: Obtain the customer history information collected by the sales end in step S2, and construct a consumer purchase feature vector x={X1, X2, ..., X j , ..., X J }, j = 1, 2, ..., J; X j is the jth consumer feature vector; S32: Calculate the mth index C by performing statistical analysis on the historical data of consumer purchase feature vectors m Relative to the nth index C n The conditional probability P(C m |C n ): Among them, P(C n |C m ) is the assumed indicator C m In important cases, indicator C n Also important is the probability, which can be obtained from historical data; P(C m ) is the indicator C m The prior probability of C is independent of other indicators. m The probability of being considered important; P(C n ) is the indicator C m The prior probability of C is that no matter how important other indicators are, n important probability; S3: Based on the calculation result of step S32, calculate the minimum possible value l of the mth indicator relative to the nth indicator mn , the most likely value k mn and the maximum possible value u mn : l mn =min(P(C m |C n ));k mn =mode(P(C m |C n ));u mn =max(P(C m |C n )); min(P(C m |C n )) is used to calculate P(C m |C n ) The function with the smallest value among the results; mode(P(C m |C n )) is used to calculate P(C m |C n ) The function of the result that appears most often in the result; max(P(C m |C n )) is used to calculate P(C m |C n )The function with the largest value among the results.

8. A cross-border e-commerce order management system as described in any one of claims 1 to 7 is used, wherein the system includes a user login mobile terminal and a central control module, characterized in that: The system also includes: a login encryption identification module, an order classification and scoring module, and an order priority sorting module; The login encryption identification module is used to obtain login user information and perform login identification on the login user using an encryption method; The order classification and scoring module is used to collect customer data from the sales end, classify different order indicators, and score each indicator of each order; The order priority sorting module is used to sort the orders according to the classification and scoring results of different order indicators based on the fuzzy hierarchical analysis method; The central control module is used to receive the recognition result of the login encryption recognition module, and then control whether to allow the logged-in user; and is used to receive the sorting order result of the order priority sorting module, and is used to control the delivery of orders in order from the most urgent to the least urgent order according to the order priority sorting order.

Images