Intelligent cross-border e-commerce order processing method and system based on block chain

Abstract

The invention discloses a block chain-based cross-border e-commerce order intelligent processing method and system, relates to the technical field of e-commerce, and aims to use a natural language processing technology for cross-border commodity orders, so that order information is subjected to accurate structured processing, manual input errors are reduced, real-time updating and sharing of order data are ensured, and the user experience is improved. According to the method, the order data are stored safely and transparently by using a block chain technology, the data are ensured not to be tampered and traceable, a warehouse is intelligently matched according to real-time inventory and transportation conditions, the situation of inventory shortage or excess is avoided, meanwhile, the path planning module calculates the optimal transportation path through an algorithm A, and the path planning efficiency is improved. And the abnormality of the path is evaluated in real time in the transportation process, so that the timeliness and reliability of transportation are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce, and particularly to an intelligent processing method and system for cross-border e-commerce orders based on blockchain. Background Art

[0002] With the advancement of globalization and the development of Internet technology, cross-border e-commerce has become an important part of global trade. Cross-border e-commerce provides consumers with convenient shopping channels through online platforms, promoting the rapid circulation of goods and trade cooperation. However, cross-border e-commerce still faces many challenges in aspects such as logistics, order processing, and inventory management. In order to improve the efficiency of the supply chain and ensure that goods can be delivered to consumers quickly and accurately, an intelligent order processing system has become a key technology in the field of cross-border e-commerce. In particular, a cross-border e-commerce order intelligent processing system combined with blockchain technology can effectively improve the reliability and efficiency of order management and logistics distribution by utilizing the decentralized, immutable, and transparent characteristics of blockchain.

[0003] In the Chinese invention patent with the publication number CN118212039A, an intelligent processing method and platform for cross-border e-commerce orders based on blockchain are disclosed. The platform collects product and seller information through a data capture module and generates a first data set, then evaluates the first data set. The user analysis module collects the user's historical order information, conducts a user reliability index evaluation, and classifies the users according to the evaluation results. The side-chain order preprocessing module matches the users with the evaluated products and sellers, sorts the orders according to the user levels, and uploads them to the main blockchain to achieve personalized order customization, improving user satisfaction and transaction efficiency, and solving the personalized needs of order allocation and processing. The main blockchain receiving and processing module and the logistics quality inspection processing module can achieve full-process monitoring and real-time detection of product quality, and effectively reduce the risks of product damage and quality defects, ensuring the safety and quality of products.

[0004] The above platform can improve user satisfaction and transaction efficiency, solve the personalized needs of order allocation and processing, and meet the diverse shopping habits and needs of users. However, in addition to this, in the existing intelligent processing methods for cross-border e-commerce orders, product listing and shipping are generally carried out manually.

[0005] However, this intelligent processing method for cross-border e-commerce orders still faces multiple bottlenecks and problems in the actual operation process. Especially in aspects such as order generation, inventory management, route planning, and handling of transportation anomalies, traditional management methods are difficult to meet the requirements of real-time, precision, and efficiency. Specifically, the generation and demand analysis of orders often rely on manual operations. Inventory matching cannot promptly reflect the actual situation of each warehouse. Route planning also lacks a dynamic adjustment mechanism for real-time traffic information and transportation anomalies. More importantly, the existing system lacks transparency and cannot monitor and respond to anomalies in the logistics transportation process in a timely manner, resulting in delays in shipping, inventory errors, and increased transportation costs, thus affecting the purchasing experience of consumers and the reputation of cross-border e-commerce platforms.

[0006] To this end, the present invention provides an intelligent processing method and system for cross-border e-commerce orders based on blockchain. Summary of the Invention

[0007] (1) Technical Problems to be Solved

[0008] In view of the deficiencies of the prior art, the present invention provides an intelligent processing method and system for cross-border e-commerce orders based on blockchain. By using natural language processing technology for cross-border commodity orders, the order information is precisely structured, reducing manual input errors and ensuring real-time update and sharing of order data. And by using blockchain technology, order data is securely and transparently stored, ensuring the immutability and traceability of data. And according to real-time inventory and transportation conditions, warehouses are intelligently matched to avoid inventory shortages or surpluses. At the same time, the route planning module calculates the optimal transportation route through the A algorithm and evaluates the anomalies of the route in real time during transportation to ensure the timeliness and reliability of transportation.

[0009] (2) Technical Solutions

[0010] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent processing system for cross-border e-commerce orders based on blockchain, including an order generation and demand analysis module, an inventory matching module, a route planning module, an anomaly handling module, and a data storage module;

[0011] The order generation and demand analysis module is used to obtain consumer order content based on the e-commerce platform, use natural language processing technology to obtain order information, and perform structured processing on the order information to obtain a structured table of order information, and upload the structured table of order information to the blockchain system;

[0012] The inventory matching module is used to obtain the real-time inventory of each warehouse based on the blockchain system, perform intelligent matching of the shipping warehouse for the commodity, and use a smart contract to lock the shipping warehouse after the shipping warehouse is matched;

[0013] The path planning module is used to build a global transportation network model and use the A algorithm to perform global path search to obtain the commodity transportation path;

[0014] The exception handling module is used to perform an exception evaluation on each segment of the commodity transportation route. If the path evaluation result is an abnormal state, the transportation calculation of this segment of the path is canceled, and the path planning is re-performed to optimize the commodity transportation;

[0015] The data storage module is used to automatically trigger the smart contract after the goods transportation is completed, collect the data related to the goods transportation, and store it in the blockchain system.

[0016] Preferably, the order generation and demand analysis module includes an information collection unit and an encryption and chaining unit;

[0017] The information collection unit is used to collect information related to product orders and inventory information, wherein the information related to product orders is obtained through the e-commerce platform where the merchant is located. After the user confirms the product order, the specific content of the product order is obtained, and the product order information is extracted using a pre-trained language model, and the extracted product order information is converted into an order information structured table, wherein the order information structured table includes order ID, product name, product quantity, product weight, destination, expected delivery time, product price, total product price, delivery method, payment method and inventory status;

[0018] Inventory information extracts commodity inventory data from the warehouse database and performs structured processing to obtain an inventory data set S;

[0019] The encryption chain unit is used to hash the order information structured table and inventory data set S using the hash encryption algorithm SHA-256 to generate a unique hash value, and upload the hash value and the encrypted order information structured table and inventory data set S to the blockchain system.

[0020] Preferably, the inventory matching module includes an inventory query unit and a warehouse matching unit;

[0021] The inventory query unit is used to obtain the commodity inventory data of each warehouse according to the blockchain system, and obtain the inventory matching score KC of each warehouse according to the commodity inventory data, wherein the commodity inventory data includes the commodity inventory quantity I available , commodity demand rate R demand , Goods transportation time T transit and the total freight cost C from the shipping warehouse to the destination transport ;

[0022] The specific method of obtaining the inventory matching score KC is as follows:

[0023]

[0024] Among them, I available represents the inventory of goods in the warehouse, and R demand represents the demand rate of goods in the warehouse, and T transit represents the goods transportation time, and C transport represents the total freight of goods from the shipping warehouse to the destination. α represents the weight coefficient of the goods demand rate R demand and β represents the weight coefficient of the transportation cost C transport .

[0025] Preferably, the warehouse matching unit is used to sort according to the inventory matching score KC of each warehouse in ascending order, and output the warehouse with the highest inventory matching score KC as the goods shipping warehouse, and trigger the intelligent contract of the blockchain system to lock the shipping warehouse.

[0026] Preferably, the path planning module includes a traffic network model construction unit and a path generation unit;

[0027] The traffic network model construction unit is used to construct a global traffic network model based on the global traffic network database and extract traffic path-related data. Among them, the global traffic network model includes traffic nodes and traffic paths;

[0028] Traffic nodes refer to ports, airports, and railway hubs;

[0029] Traffic paths refer to the traffic paths between two adjacent traffic nodes;

[0030] Traffic path-related data refers to the goods transportation time and unit transportation cost;

[0031] According to the traffic path-related data, feature extraction is performed to obtain the path transportation reliability score YS. Among them, the method for obtaining the path transportation reliability score YS is:

[0032]

[0033] In the formula, P ontime represents the on-time rate of goods transportation on the traffic path, that is, the ratio of the number of on-time transports on the traffic path to the total number of transports, and T delay represents the average delay time of goods on the traffic path, and T max represents the maximum delay time of goods on the traffic path, and N delayed represents the number of goods delay times on the traffic path, and N transports represents the total number of transports of goods on the traffic path, and ω1, ω2, and ω3 are weight coefficients;

[0034] Preset a transportation reliability score threshold YSL, and conduct a comparative analysis on the transportation paths between adjacent nodes in the global traffic network model using the preset transportation reliability score threshold YSL and the path transportation reliability score YS to evaluate the feasibility of the transportation paths. The specific evaluation process is as follows:

[0035] If the path transportation reliability score YS is greater than or equal to the transportation reliability score threshold YSL, that is, YS≥YSL, then it is determined that the feasibility of the transportation path is abnormal. At this time, mark the path as an infeasible path and remove the path in the global traffic network model;

[0036] If the path transportation reliability score YS is less than the transportation reliability score threshold YSL, that is, YS<YSL, then it is determined that the feasibility of the transportation path is normal and no processing is required;

[0037] Update the global traffic network model according to the path feasibility evaluation results.

[0038] Preferably, the path generation unit is used to search for paths using the A algorithm based on the global traffic network model to obtain the optimal path for commodity transportation. Among them, the specific process of path search is as follows:

[0039] Set the shipping warehouse as the initial node and the destination as the end node, and load the updated global traffic network model. Extract all traffic nodes from the initial node to the end node, construct a candidate traffic node set H, and construct a heuristic function h(n) and an actual cost function g(n). Among them, the specific form of the heuristic function h(n) is:

[0040]

[0041] Among them, (x1, y1) and (x2, y2) respectively represent the coordinates of two adjacent traffic nodes;

[0042] The specific form of the actual cost function g(n) is:

[0043]

[0044] In the formula, C mode represents the unit transportation cost of the commodity transportation method between two adjacent traffic nodes, T mode represents the estimated transportation time of the commodity transportation method between two adjacent traffic nodes, and respectively represent the unit transportation cost C mode of the commodity transportation method between two adjacent traffic nodes and the estimated transportation time T mode of the commodity transportation method between two adjacent traffic nodes. Among them, the commodity transportation method refers to air transportation, railway transportation, ship transportation, and automobile transportation existing between two adjacent traffic nodes;

[0045] Taking the initial node as the traffic node where the current commodity is located, according to the heuristic function h(n) and the actual cost function g(n), obtain the total cost function f(n) of all traffic nodes directly existing from the initial node to the next target node. Among them, the way to obtain the total cost function f(n) is:

[0046] f(n) = h(n) + g(n);

[0047] According to the total cost function f(n), select the traffic node with the smallest total cost function f(n) and add it to the closed node list, where the closed node list is used to store the node with the smallest total cost function f(n);

[0048] Update the traffic node with the smallest total cost function f(n) as the traffic node where the current commodity is located, and conduct the search for the next traffic node. Repeat the calculation of the total cost function f(n) and the update process of the traffic node until all traffic nodes from the initial node to the terminal node are obtained, construct the commodity transportation path, and output the commodity transportation path at this time as the optimal commodity transportation path.

[0049] Preferably, the exception handling module includes a transportation path monitoring unit and an exception evaluation unit;

[0050] The transportation path monitoring unit is used to obtain N sections of traffic paths according to the optimal commodity transportation path, and use Internet of Things devices to collect in real time the data related to commodity transportation at each traffic node in the optimal commodity transportation path, and upload the data related to commodity transportation to the blockchain system, where the data related to commodity transportation includes the commodity arrival time, the commodity departure time, the traffic flow, and the historical traffic data;

[0051] According to the data related to commodity transportation, perform feature extraction to obtain the time deviation coefficient SJ and the traffic jam coefficient JT of each section of traffic path;

[0052] The way to obtain the time deviation coefficient SJ is:

[0053]

[0054] In the formula, represents the expected transportation time of the commodity when it is on the j-th section of traffic path, represents the shortest transportation time of the commodity when it is on the j-th section of traffic path, j = [1, 2, 3,..., N];

[0055] The way to obtain the traffic jam coefficient JT is:

[0056]

[0057] In the formula, N represents the total number of traffic paths in the optimal commodity transportation path, Tideal，j Denote the shortest travel time of the commodity on the j-th section of the traffic path as T actual，j Denote the expected travel time of the commodity on the j-th section of the traffic path.

[0058] Preferably, the anomaly evaluation unit is used to perform a summary calculation based on the obtained time deviation coefficient SJ, traffic congestion coefficient JT, and combined with the path transportation reliability score YS to obtain the path anomaly score PF for each traffic path. Among them, the method for obtaining the path anomaly score PF is as follows:

[0059] PF = ω1·SJ + ω2·JT + ω3·YS + C;

[0060] In the formula, ω1, ω2, and ω3 respectively represent the weight coefficients of the time deviation coefficient SJ, traffic congestion coefficient JT, and path transportation reliability score YS, and C represents the first correction constant;

[0061] Preset the path anomaly score threshold PFYZ, and compare and analyze the path anomaly score PF and the path anomaly score threshold PFYZ to evaluate the path anomaly state. The specific evaluation content is as follows:

[0062] If the path anomaly score PF is greater than or equal to the path anomaly score threshold PFYZ, it is determined that the transportation path is in an abnormal state. At this time, cancel the transportation plan for this section of the path, and re-run the path generation unit to modify the transportation path;

[0063] If the path anomaly score PF is less than the path anomaly score threshold PFYZ, it is determined that the transportation path is in a normal state and no intervention is required.

[0064] Preferably, the data storage module is used to trigger the smart contract of the blockchain system after the commodity arrives at the destination, automatically update the order status, complete the delivery, store the order data, transportation path, and anomaly handling information on the blockchain, and update the commodity inventory data after the order is completed.

[0065] Preferably, a cross-border e-commerce order intelligent processing method based on blockchain includes the following steps

[0066] Step 1: Obtain the consumer order content based on the e-commerce platform, use natural language processing technology to obtain the order information, perform structured processing on the order information to obtain the structured order information table, and upload the structured order information table to the blockchain system;

[0067] Step 2: Obtain the real-time inventory of each warehouse based on the blockchain system, perform intelligent matching of the commodity delivery warehouse, and after matching the delivery warehouse, use the smart contract to lock the delivery warehouse;

[0068] Step 3: Construct a global transportation network model and use the A algorithm for global path search to obtain the commodity transportation path;

[0069] Step 4: For each section of the commodity transportation path, conduct an anomaly assessment. If the path assessment result is in an abnormal state, cancel the transportation calculation for this section of the path and re - conduct path planning to optimize commodity transportation;

[0070] Step 5: After the commodity transportation is completed, automatically trigger a smart contract, collect data related to commodity transportation, and store it in the blockchain system.

[0071] The present invention provides a method and system for intelligent processing of cross - border e - commerce orders based on blockchain, having the following beneficial effects:

[0072] (1) By structuring order information and uploading it to the blockchain, the system can achieve real - time sharing and transparent tracking of order information, avoiding errors and delays that may be caused by manual input, human intervention, etc. in the traditional process. In addition, through smart contracts to lock the shipping warehouse, path planning, and anomaly handling, the system realizes real - time optimization of warehouse selection and transportation paths, ensuring the timeliness and accuracy of order processing. The overall system can improve the efficiency from order generation to delivery, providing faster and more accurate services for cross - border e - commerce platforms.

[0073] (2) Through the intelligent warehouse matching and path planning modules, the logistics cost is effectively reduced. In terms of warehouse matching, the system combines real - time inventory and transportation costs to intelligently select the most suitable warehouse for shipping, thereby reducing unnecessary inventory backlogs and transportation waste. The path planning module uses the A algorithm combined with global transportation network data to accurately calculate the optimal transportation path and dynamically adjust according to factors such as real - time traffic flow and transportation time, effectively avoiding problems such as traffic congestion and delays. Through these measures, the system greatly optimizes the allocation of logistics resources, improves transportation efficiency, and reduces transportation costs.

[0074] (3) Based on blockchain technology, the intelligent processing system for cross - border e - commerce orders makes each link in the commodity transportation process highly transparent and traceable. All order data, inventory information, transportation paths, and anomaly handling records are recorded on the blockchain in real - time, ensuring that the data cannot be tampered with and can be checked at any time. Consumers and platforms can view the order status in real - time to ensure that any problems during transportation can be discovered and processed in a timely manner. In addition, through the anomaly handling module, the system can predict potential risks during transportation and automatically adjust the path, reducing uncontrollable factors during transportation and improving customer satisfaction. Description of the Drawings

[0075] Figure 1Block diagram of an intelligent processing system for cross-border e-commerce orders based on blockchain according to the present invention.

[0076] Figure 2 Schematic flow diagram of a method for intelligent processing of cross-border e-commerce orders based on blockchain according to the present invention.

[0077] Figure 3 Schematic flow diagram of the inventory matching module according to the present invention.

[0078] Figure 4 Schematic flow diagram of the path planning module according to the present invention.

[0079] Figure 5 Schematic diagram of the total freight for a commodity from shipment to the destination. Specific implementation manners

[0080] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0081] Embodiment 1

[0082] Please refer to Figure 1 、 Figure 3 、 Figure 4 and Figure 5 , the present invention provides an intelligent processing system for cross-border e-commerce orders based on blockchain, including an order generation and demand analysis module, an inventory matching module, a path planning module, an exception handling module, and a data storage module;

[0083] The order generation and demand analysis module is used to obtain the content of consumer orders based on the e-commerce platform, and use natural language processing technology to obtain order information, and perform structured processing on the order information to obtain a structured table of order information, and upload the structured table of order information to the blockchain system;

[0084] The inventory matching module is used to obtain the real-time inventory of each warehouse based on the blockchain system, and perform intelligent matching of the shipping warehouses for the commodities, and after matching the shipping warehouse, use a smart contract to lock the shipping warehouse;

[0085] The path planning module is used to construct a global transportation network model and use the A algorithm to perform global path search to obtain the commodity transportation path;

[0086] The exception handling module is used to perform an exception evaluation on each segment of the commodity transportation route. If the path evaluation result is an abnormal state, the transportation calculation of this segment of the path is canceled, and the path planning is re-performed to optimize the commodity transportation;

[0087] The data storage module is used to automatically trigger the smart contract after the goods transportation is completed, collect the data related to the goods transportation, and store it in the blockchain system.

[0088] In the embodiment, through the close cooperation of each module, the efficiency and accuracy of order processing are improved. First, the order generation and demand analysis module automatically obtains and structures order information from the e-commerce platform through natural language processing technology, ensuring the efficient extraction and accurate transmission of order data, and uploading it to the blockchain system, realizing the transparency and immutability of order information, and providing a reliable data basis for subsequent links. Secondly, the inventory matching module uses real-time inventory information and smart contracts to lock the shipping warehouse, optimizes warehouse selection and inventory management, reduces the problems of insufficient inventory and unreasonable allocation, thereby improving logistics efficiency and accuracy. The path planning module searches for paths through the global transportation network model and A algorithm, and can obtain the optimal transportation path in real time, avoiding path delays and transportation cost waste in traditional path planning, and effectively improving transportation efficiency. Further, the exception handling module can detect potential problems in real time and adjust the path through abnormal evaluation of each section of the path during transportation, ensuring the punctuality and safety of commodity transportation. Finally, the data storage module ensures that the entire process information of commodity transportation is recorded and stored on the blockchain, improving the traceability of the system, facilitating subsequent inquiries and audits, and improving transparency and customer trust.

[0089] Example 2

[0090] Please refer to Figure 1 ,Specifically: the order generation and demand analysis module includes an ,information collection unit and an encryption and chain unit;

[0091] The information collection unit is used to collect information related to product orders and inventory information, wherein the information related to product orders is obtained through the e-commerce platform where the merchant is located. After the user confirms the product order, the specific content of the product order is obtained, and the product order information is extracted using a pre-trained language model, and the extracted product order information is converted into an order information structured table, wherein the order information structured table includes order ID, product name, product quantity, product weight, destination, expected delivery time, product price, total product price, delivery method, payment method and inventory status;

[0092] Specific example:

[0093] Suppose there is a cross-border e-commerce order. The order information is extracted and structured to obtain a structured order information table. The structured order information table is shown in Table 1 below:

[0094] Field Description Example Data Order ID Unique identifier for each order ORD123456789 Product Name Name of the product Smartphone Quantity of Goods Quantity of the product 2 Product Weight Weight of the product 1.5 kg Destination Delivery address Beijing, China Expected Delivery Time Delivery time requested by the consumer October 1, 2024 Product Price Unit price of the product ￥999 Total Price of Goods Total price of the product ￥1998 Delivery Method Selected delivery method SF Express Payment Method Method of paying for the order Alipay Inventory Status Whether the product is in stock In stock

[0095] Table 1

[0096] The inventory information extracts the commodity inventory data through the warehouse database and structures it to obtain the inventory data set S;

[0097] The encryption and blockchain unit is used to hash-encrypt the structured order information table and the inventory data set S using the hash encryption algorithm SHA-256 to generate a unique hash value, and upload the hash value and the encrypted structured order information table and inventory data set S to the blockchain system.

[0098] In the embodiment, through intelligent information collection and encryption and blockchain processing, the efficiency and accuracy of cross-border e-commerce order processing can be improved. First, the information collection unit obtains the consumer's order information in real time through seamless docking with the e-commerce platform, and uses a pre-trained language model to automatically extract the order content. By automatically extracting key information such as commodity name, quantity, weight, destination, and expected delivery time, errors and delays that may occur in the traditional manual input process are avoided, ensuring the accuracy of the order information. At the same time, the inventory information is also obtained in real time through the warehouse database. Combined with the structuring of the inventory data set S, the efficiency of inventory matching is further improved, which enables the real-time update of the warehouse inventory situation, avoids problems such as insufficient inventory or lagging inventory information, and improves the intelligence level of the shipping warehouse selection. The effective combination of inventory and order information ensures the accuracy of warehouse matching, thus providing reliable data support for subsequent path planning and order shipping. The encryption and blockchain unit encrypts the order information and inventory data using the hash encryption algorithm SHA-256, generates a unique hash value and uploads it to the blockchain system, ensuring the immutability and transparency of the data. The blockchain technology ensures the security of data storage and provides a real-time query function, enabling any party to conduct data verification when needed. This not only improves the reliability of order processing but also increases the transparency of the system and the trust of users.

[0099] Embodiment 3

[0100] Please refer to Figure 1 、 Figure 3 and Figure 5 Specifically: The inventory matching module includes an inventory query unit and a warehouse matching unit;

[0101] The inventory query unit is used to obtain the commodity inventory data of each warehouse based on the blockchain system, and obtain the inventory matching score KC of each warehouse according to the commodity inventory data, where the commodity inventory data includes the commodity inventory quantity I available , the commodity demand rate R demand , the commodity transportation time T transit and the total freight C of the commodity from the shipping warehouse to the destination transport ;

[0102] The specific method for obtaining the inventory matching score KC is as follows:

[0103]

[0104] Among them, I available represents the commodity inventory quantity of the warehouse, R demand represents the commodity demand rate of the warehouse, T transit represents the commodity transportation time, C transport represents the total freight of the commodity from the shipping warehouse to the destination, α represents the weight coefficient of the commodity demand rate R demand and β represents the weight coefficient of the transportation cost C transport . Among them, the specific data of the weight coefficient is set by the customer according to the actual situation, 0 < α < 1, 0 < β < 1, and α + β = 1.

[0105] The commodity transportation time T transit is obtained through the transportation management system;

[0106] The commodity demand rate R demand is obtained through historical sales data;

[0107] The total freight C of the commodity from the shipping warehouse to the destination transport is obtained through the contract agreement of the logistics service provider;

[0108] The warehouse matching unit is used to sort according to the inventory matching score KC of each warehouse in ascending order, and output the warehouse with the highest inventory matching score KC as the commodity shipping warehouse, and trigger the smart contract of the blockchain system to lock the shipping warehouse.

[0109] Specific example:

[0110] Suppose in the automatic processing of cross-border e-commerce orders, the commodities are stored in Warehouse No. 1, Warehouse No. 2, and Warehouse No. 3 respectively. After receiving the commodity order, inventory matching is required to select a warehouse from Warehouse No. 1, Warehouse No. 2, and Warehouse No. 3 for commodity shipping, as shown in Table 2 below:

[0111]

[0112] Table 2

[0113] Calculate the inventory matching scores KC1, KC2, and KC3 of Warehouse 1, Warehouse 2, and Warehouse 3 respectively:

[0114]

[0115]

[0116] Among them, if KC3 > KC1 > KC2, then the inventory matching score of Warehouse 3 is the highest. Select Warehouse 3 as the shipping warehouse and lock the shipping warehouse.

[0117] In the embodiment, through the inventory query unit, the system can obtain the commodity inventory data of each warehouse in real time based on blockchain technology, ensuring the transparency and accuracy of inventory information. This module not only depends on the static inventory quantity but also combines factors such as commodity demand rate, transportation time, and transportation cost to evaluate the inventory matching score KC of each warehouse. This comprehensive score provides a scientific basis for warehouse selection. By adjusting the weight coefficients of commodity demand rate and transportation cost, the system can flexibly optimize inventory allocation and logistics arrangements according to business needs. The warehouse matching unit automatically selects the warehouse with the highest inventory matching score as the shipping warehouse according to the ranking of the inventory matching score KC. This process reduces manual intervention and improves the automation degree of order processing. At the same time, the introduction of smart contracts ensures the effective execution of the inventory matching results, and the warehouse selection and inventory locking operations cannot be tampered with, enhancing the trust and security of the system. This blockchain-based real-time inventory management not only ensures the accurate matching of warehouse inventory and order requirements but also optimizes the allocation of warehouse resources, avoiding the phenomenon of overstock or shortage.

[0118] Embodiment 4

[0119] Please refer to Figure 1 、 Figure 4 and Figure 5 , specifically: The path planning module includes a traffic network model construction unit and a path generation unit;

[0120] The traffic network model construction unit is used to construct a global traffic network model based on the global traffic network database and extract traffic path-related data. Among them, the global traffic network model includes traffic nodes and traffic paths;

[0121] Traffic nodes refer to ports, airports, and railway hubs;

[0122] Traffic paths refer to the traffic paths between two adjacent traffic nodes;

[0123] Traffic path-related data refers to commodity transportation time and unit transportation cost;

[0124] Based on traffic path - related data, feature extraction is performed to obtain the path transportation reliability score YS. Among them, the method for obtaining the path transportation reliability score YS is as follows:

[0125]

[0126] In the formula, P ontime represents the on - time rate of commodity transportation on the traffic path, that is, the ratio of the number of on - time transports on the traffic path to the total number of transports, T delay represents the average delay time of commodities on the traffic path, T max represents the maximum delay time of commodities on the traffic path, N delayed represents the number of commodity delay times on the traffic path, N transports represents the total number of transports of commodities on the traffic path. ω1, ω2, and ω3 are weight coefficients, and the specific values of the weight coefficients are obtained by the customer according to the actual situation. 0 < ω1 < 1, 0 < ω2 < 1, 0 < ω3 < 1, and ω1 + ω2 + ω3 = 1;

[0127] A preset transportation reliability score threshold YSL is set, and for the traffic paths between adjacent nodes in the global traffic network model, a comparative analysis is carried out using the preset transportation reliability score threshold YSL and the path transportation reliability score YS to evaluate the feasibility of the traffic path. The specific evaluation process is as follows:

[0128] If the path transportation reliability score YS is greater than or equal to the transportation reliability score threshold YSL, that is, YS ≥ YSL, then it is determined that the feasibility of the traffic path is abnormal. At this time, the path is marked as an unfeasible path and removed from the global traffic network model;

[0129] If the path transportation reliability score YS is less than the transportation reliability score threshold YSL, that is, YS < YSL, then it is determined that the feasibility of the traffic path is normal and no processing is required;

[0130] Based on the path feasibility evaluation results, the global traffic network model is updated.

[0131] The path generation unit is used to perform path search using the A algorithm based on the global traffic network model to obtain the optimal path for commodity transportation;

[0132] The A algorithm is an algorithm applied to path search and graph traversal, and is often used in scenarios such as maps, games, and robot navigation. The A algorithm is based on heuristic search and combines the advantages of breadth - first search (BFS) and greedy best - first search. It can find the optimal path from the starting point to the target point.

[0133] Among them, the specific process of path search is as follows:

[0134] Set the shipping warehouse as the initial node, the destination as the terminal node, load the updated global traffic network model, extract all traffic nodes from the initial node to the terminal node, construct a set H of candidate traffic nodes, and construct a heuristic function h(n) and an actual cost function g(n). Among them, the specific form of the heuristic function h(n) is:

[0135]

[0136] Among them, (x1, y1) and (x2, y2) respectively represent the coordinates of two adjacent traffic nodes;

[0137] The specific form of the actual cost function g(n) is:

[0138]

[0139] In the formula, C mode represents the unit transportation cost of the commodity transportation mode between two adjacent traffic nodes, T mode represents the estimated transportation time of the commodity transportation mode between two adjacent traffic nodes, and respectively represent the unit transportation cost C mode of the commodity transportation mode between two adjacent traffic nodes and the estimated transportation time T mode of the commodity transportation mode between two adjacent traffic nodes. The weight coefficients of them, where the commodity transportation mode refers to air transportation, railway transportation, ship transportation and road transportation existing between two adjacent traffic nodes. The specific data of the weight coefficients are set by the customer according to the actual situation. And

[0140] Taking the initial node as the current traffic node where the commodity is located, according to the heuristic function h(n) and the actual cost function g(n), obtain the total cost function f(n) of all traffic nodes directly existing from the initial node to the next target node. Among them, the way to obtain the total cost function f(n) is:

[0141] f(n) = h(n) + g(n);

[0142] According to the total cost function f(n), select the traffic node with the smallest total cost function f(n) and add it to the closed node list, where the closed node list is used to store the node with the smallest total cost function f(n);

[0143] Update the traffic node with the minimum total cost function f(n) to the traffic node where the current commodity is located, and conduct the search for the next traffic node. Repeat the calculation of the total cost function f(n) and the process of traffic node update until all traffic nodes from the initial node to the terminal node are obtained, construct the commodity transportation path, and output the commodity transportation path at this time as the optimal commodity transportation path.

[0144] In the embodiment, through the collaborative work of the traffic network model construction unit and the path generation unit, the selection efficiency and accuracy of the commodity transportation path are improved. The traffic network model construction unit extracts the relevant information of key traffic nodes and paths, including transportation time and cost, from the real-time global traffic network data, thus providing strong data support for subsequent path optimization. By calculating the path transportation reliability score YS, the system can evaluate the feasibility of each traffic path based on factors such as on-time rate and delay time to ensure the reliability of the selected transportation path. In addition, the path generation unit adopts the A algorithm to search the global traffic nodes using the heuristic function and the actual cost function to minimize the transportation cost and time, ensuring that the selected path is optimal in terms of both cost and time. Through the calculation of the total cost function and node update, the system can dynamically select the optimal path, effectively avoiding the inefficiency and obsolescence problems in traditional path planning methods. The system's real-time monitoring of path feasibility and abnormal evaluation further optimize the transportation process and reduce the delay risk caused by traffic jams or emergencies.

[0145] Embodiment 5

[0146] Please refer to Figure 1 and Figure 5 , specifically: The exception handling module includes a transportation path monitoring unit and an abnormal evaluation unit;

[0147] The transportation path monitoring unit is used to obtain N sections of traffic paths according to the optimal commodity transportation path, and use Internet of Things devices to collect the commodity transportation-related data of each traffic node in the optimal commodity transportation path in real time, and upload the commodity transportation-related data to the blockchain system. Among them, the commodity transportation-related data includes the commodity arrival time, the commodity departure time, the traffic flow, and the historical traffic data;

[0148] According to the commodity transportation-related data, feature extraction is performed to obtain the time deviation coefficient SJ and the traffic jam coefficient JT of each section of traffic path;

[0149] The way to obtain the time deviation coefficient SJ is:

[0150]

[0151] In the formula, represents the expected transportation time of the commodity when it is on the j-th section of traffic path, Denote the shortest transportation time of the commodity when it is on the j-th section of the transportation path, where j = [1, 2, 3, …, N];

[0152] The expected transportation time of the commodity when it is on the j-th section of the transportation path Obtained through the traffic prediction system;

[0153] The shortest transportation time of the commodity when it is on the j-th section of the transportation path is obtained through historical traffic data;

[0154] The way to obtain the traffic congestion coefficient JT is:

[0155]

[0156] In the formula, N represents the total number of traffic paths in the optimal commodity transportation path, T ideal，j Denote the shortest passing time of the commodity on the j-th section of the transportation path, T actual，j Denote the expected passing time of the commodity on the j-th section of the transportation path.

[0157] The shortest passing time T of the commodity on the j-th section of the transportation path ideal，j Obtained through historical traffic data;

[0158] The expected passing time T of the commodity on the j-th section of the transportation path actual，j Obtained through the traffic prediction system;

[0159] Among them, the transportation time includes the driving time of the commodity on the path and the staying time at traffic nodes, and the passing time includes the driving time of the commodity on the path;

[0160] The abnormal evaluation unit is used to perform summary calculations based on the obtained time deviation coefficient SJ, traffic congestion coefficient JT, and combined with the path transportation reliability score YS to obtain the path abnormal score PF of each traffic path. Among them, the way to obtain the path abnormal score PF is:

[0161] PF = ω1·SJ + ω2·JT + ω3·YS + C;

[0162] In the formula, ω1, ω2, and ω3 respectively represent the weight coefficients of the time deviation coefficient SJ, traffic congestion coefficient JT, and path transportation reliability score YS, and C represents the first correction constant;

[0163] Preset the path abnormal score threshold PFYZ, and compare and analyze the path abnormal score PF and the path abnormal score threshold PFYZ to evaluate the path abnormal state. The specific evaluation content is as follows:

[0164] If the path abnormal score PF is greater than or equal to the path abnormal score threshold PFYZ, it is determined that the transportation path is in an abnormal state. At this time, cancel the transportation plan for this section of the path, and re-run the path generation unit to modify the transportation path;

[0165] If the path anomaly score PF is less than the path anomaly score threshold PFYZ, it is determined that the transportation path is in a normal state and no intervention is required.

[0166] In the embodiment, by integrating the transportation path monitoring unit and the anomaly evaluation unit, the intelligent monitoring and dynamic adjustment capabilities of the transportation path are improved. First, the transportation path monitoring unit uses Internet of Things devices to collect real-time data related to the transportation of goods at each traffic node, including the arrival time, departure time, traffic flow, and historical traffic data of the goods. These information can timely reflect potential problems in the transportation process and provide accurate real-time data support. By calculating the time deviation coefficient SJ and the traffic jam coefficient JT, the system can identify the differences between the actual transportation performance and the expected performance of each path. Further, in combination with the path transportation reliability score YS, the anomaly evaluation unit can comprehensively evaluate each path, calculate the path anomaly score PF of each path, and compare it with the preset path anomaly score threshold PFYZ to determine whether there is an anomaly in the path. This mechanism effectively predicts potential transportation risks, can timely detect and handle transportation paths that may cause delays or cost increases, and avoids unnecessary losses.

[0167] Embodiment 6

[0168] Please refer to Figure 1 , specifically: The data storage module is used to trigger the smart contract of the blockchain system after the goods arrive at the destination, automatically update the order status, complete the delivery, and store the order data, transportation path, and anomaly handling information on the blockchain, and update the commodity inventory data after the order is completed.

[0169] In the embodiment, after the goods arrive at the destination, the smart contract of the blockchain is automatically triggered to update the order status and complete the delivery in real time. The delivery and status update of the order no longer rely on manual operations, reducing delays and errors caused by human intervention, and improving the efficiency and accuracy of order processing. In addition, after the order is completed, the data storage module will also automatically update the commodity inventory data. This function helps warehouse managers to master the inventory situation in real time and avoid problems such as insufficient inventory or out-of-stock. Through the blockchain system, the update of inventory data can be synchronized to the entire supply chain in real time, ensuring the consistency and accuracy of data. In this way, the entire cross-border e-commerce system can respond more flexibly to market demands, improving the reaction speed and flexibility of the supply chain.

[0170] Embodiment 7

[0171] Please refer to Figure 2 , specifically: A method for intelligent processing of cross-border e-commerce orders based on blockchain, including the following steps,

[0172] Step 1: Based on the e-commerce platform, obtain the consumer order content, use natural language processing technology to obtain the order information, perform structured processing on the order information to obtain a structured table of order information, and upload the structured table of order information to the blockchain system;

[0173] Step 2: Based on the blockchain system, obtain the real-time inventory of each warehouse, perform intelligent matching of the shipping warehouse for the goods, and after matching the shipping warehouse, use a smart contract to lock the shipping warehouse;

[0174] Step 3: Construct a global transportation network model and use the A algorithm for global path search to obtain the goods transportation path;

[0175] Step 4: For each section of the goods transportation path, conduct an anomaly assessment. If the path assessment result is in an abnormal state, cancel the transportation calculation for that section of the path and re-plan the path to optimize the goods transportation;

[0176] Step 5: After the goods transportation is completed, automatically trigger the smart contract to collect the data related to the goods transportation and store it in the blockchain system.

[0177] In the embodiment, through the full-process automation and intelligence of cross-border commodity orders, the efficiency of cross-border e-commerce order processing is improved. Through natural language processing technology, the order information is accurately structured, reducing manual input errors and ensuring the real-time update and sharing of order data. Using blockchain technology, all order data is stored securely and transparently, ensuring the immutability and traceability of the data. In terms of warehouse matching, the system intelligently matches the warehouse according to the real-time inventory and transportation conditions and locks the shipping warehouse through a smart contract, avoiding the situation of inventory shortage or excess. At the same time, the path planning module calculates the optimal transportation path through the A algorithm and evaluates the anomalies of the path in real time during transportation to ensure the timeliness and reliability of transportation. After the goods transportation is completed, the system automatically collects and stores the transportation data through the smart contract, providing full-process transparent and real-time monitoring. This series of measures improves the logistics efficiency, reduces the logistics cost, enhances the trust of consumers, and improves the competitiveness of cross-border e-commerce platforms.

[0178] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cross-border e-commerce order intelligent processing system based on blockchain, characterized by: It includes order generation and demand analysis module, inventory matching module, route planning module, exception handling module and data storage module; The order generation and demand analysis module is used to obtain the content of consumer orders based on the e-commerce platform, and use natural language processing technology to obtain order information, and perform structured processing on the order information, obtain the structured form of order information, and upload the structured form of order information to the blockchain system; The inventory matching module is used to obtain the real-time inventory of each warehouse based on the blockchain system, and to intelligently match the product delivery warehouse. After matching the delivery warehouse, the delivery warehouse is locked using a smart contract. The path planning module is used to build a global transportation network model and use the A algorithm to perform global path search to obtain the commodity transportation path; The exception handling module is used to perform an exception evaluation on each segment of the commodity transportation route. If the path evaluation result is an abnormal state, the transportation calculation of this segment of the path is canceled, and the path planning is re-performed to optimize the commodity transportation; The data storage module is used to automatically trigger the smart contract after the goods are transported, collect the data related to the goods transportation, and store it in the blockchain system.

2. A cross-border e-commerce order intelligent processing system based on blockchain according to claim 1, characterized in that: The order generation and demand analysis module includes an information collection unit and an encryption and chaining unit; The information collection unit is used to collect information related to product orders and inventory information, wherein the information related to product orders is obtained through the e-commerce platform where the merchant is located. After the user confirms the product order, the specific content of the product order is obtained, and the product order information is extracted using a pre-trained language model, and the extracted product order information is converted into an order information structured table, wherein the order information structured table includes order ID, product name, product quantity, product weight, destination, expected delivery time, product price, total product price, delivery method, payment method and inventory status; Inventory information extracts commodity inventory data from the warehouse database and performs structured processing to obtain an inventory data set S; The encryption chain unit is used to hash the order information structured table and inventory data set S using the hash encryption algorithm SHA-256 to generate a unique hash value, and upload the hash value and the encrypted order information structured table and inventory data set S to the blockchain system.

3. A blockchain-based cross-border e-commerce order intelligent processing system according to claim 2, characterized in that: The inventory matching module includes an inventory query unit and a warehouse matching unit; The inventory query unit is used to obtain the commodity inventory data of each warehouse according to the blockchain system, and obtain the inventory matching score KC of each warehouse according to the commodity inventory data, wherein the commodity inventory data includes the commodity inventory quantity I available , commodity demand rate R demand , Goods transportation time T transit and the total freight cost C from the shipping warehouse to the destination transport ; The specific method of obtaining the inventory matching score KC is as follows: Among them, I available Represents the inventory of goods in the warehouse, R demand represents the demand rate of goods in the warehouse, T transit Indicates the transportation time of goods, C transport represents the total freight cost of the product from the shipping warehouse to the destination, and α represents the product demand rate R demand The weight coefficient, β represents the transportation cost C transport The weight coefficient of .

4. A cross-border e-commerce order intelligent processing system based on blockchain according to claim 3, characterized in that: The warehouse matching unit is used to sort the inventory matching scores KC of each warehouse from low to high, and output the warehouse with the highest inventory matching score KC as the product shipping warehouse, and trigger the blockchain system smart contract to lock the shipping warehouse.

5. A blockchain-based cross-border e-commerce order intelligent processing system according to claim 4, characterized in that: The path planning module includes a traffic network model building unit and a path generation unit; The traffic network model building unit is used to build a global traffic network model based on the global traffic network database and extract traffic path related data, wherein the global traffic network model includes traffic nodes and traffic paths; Transportation nodes refer to ports, airports, and railway hubs; Traffic path refers to the traffic path between two adjacent traffic nodes; Data related to transportation routes refer to the transportation time and unit transportation cost of goods; Based on the traffic path related data, feature extraction is performed to obtain the path transportation reliability score YS, where the path transportation reliability score YS is obtained as follows: Where P ontime It represents the on-time rate of commodity transportation along the transportation path, that is, the ratio of on-time transportation times along the transportation path to the total transportation times. delay represents the average delay time of the goods in the traffic path, T max Indicates the maximum product delay time of the traffic path, N delayed Indicates the number of product delays along the transportation path, N transports represents the total number of transportation times of the goods on the transportation path, and ω1, ω2 and ω3 are weight coefficients; The transport reliability score threshold YSL is preset, and the transport paths between adjacent nodes in the global transport network model are compared and analyzed using the preset transport reliability score threshold YSL and the path transport reliability score YS to evaluate the feasibility of the transport path. The specific evaluation process is as follows: If the path transport reliability score YS is greater than or equal to the transport reliability score threshold YSL, that is, YS ≥ YSL, the feasibility of the transportation path is determined to be abnormal. At this time, the path is marked as an infeasible path and is removed from the global transportation network model; If the path transport reliability score YS is less than the transport reliability score threshold YSL, that is, YS<YSL, then the feasibility of the transportation path is determined to be normal and no processing is required; Update the global transportation network model based on the path feasibility assessment results.

6. A cross-border e-commerce order intelligent processing system based on blockchain according to claim 5, characterized in that: The path generation unit is used to search for paths using the A algorithm based on the global transportation network model to obtain the optimal path for commodity transportation. The specific process of path search is as follows: Set the shipping warehouse as the initial node and the destination as the final node, load the updated global transportation network model, extract all transportation nodes from the initial node to the final node, construct the candidate transportation node set H, and construct the heuristic function h(n) and the actual cost function g(n), where the heuristic function h(n) is specifically expressed as: Among them, (x1, y1) and (x2, y2) represent the coordinates of two adjacent traffic nodes respectively; The actual cost function g(n) is specifically expressed as: In the formula, C mode It represents the unit transportation cost of the commodity transportation mode between two adjacent transportation nodes, T mode Indicates the estimated transportation time of the commodity transportation mode between two adjacent transportation nodes. and Represents the unit transportation cost C of the commodity transportation mode between two adjacent transportation nodes mode The estimated transportation time T of the commodity transportation mode between two adjacent transportation nodes mode The weight coefficient of , where the commodity transportation mode refers to the air transportation, railway transportation, ship transportation and automobile transportation between two adjacent transportation nodes; Take the initial node as the node where the current product is located, and obtain the total cost function f(n) of all traffic nodes directly from the initial node to the next target node based on the heuristic function h(n) and the actual cost function g(n). The total cost function f(n) is obtained as follows: f(n)=h(n)+g(n); According to the total cost function f(n), select the traffic node with the smallest total cost function f(n) and add it to the closed node list, where the closed node list is used to store the node with the smallest total cost function f(n); Update the traffic node with the smallest total cost function f(n) as the traffic node where the current product is located, and search for the next traffic node. Repeat the total cost function f(n) calculation and traffic node update process until all traffic nodes from the initial node to the final node are obtained, build the product transportation path, and output the current product transportation path as the optimal product transportation path.

7. A cross-border e-commerce order intelligent processing system based on blockchain according to claim 6, characterized in that: The exception handling module includes a transport path monitoring unit and an exception assessment unit; The transportation path monitoring unit is used to obtain N traffic paths according to the optimal commodity transportation path, and use the Internet of Things device to collect commodity transportation related data of each traffic node in the optimal commodity transportation path in real time, and upload the commodity transportation related data to the blockchain system, where the commodity transportation related data includes commodity arrival time, commodity departure time, traffic flow and historical traffic data; Based on the relevant data of commodity transportation, feature extraction is performed to obtain the time deviation coefficient SJ and traffic congestion coefficient JT of each traffic path; The time deviation coefficient SJ is obtained as follows: In the formula, represents the expected transportation time of the product on the jth transportation route, represents the shortest transportation time of the commodity on the jth transportation route, j = [1, 2, 3, ..., N]; The traffic congestion coefficient JT is obtained as follows: Where N represents the total number of transportation paths in the optimal commodity transportation path, T ideal，j represents the shortest travel time of the product in the jth section of the traffic path, T actual，j It represents the expected travel time of the commodity in the jth traffic path.

8. A blockchain-based cross-border e-commerce order intelligent processing system according to claim 7, characterized in that: The abnormality assessment unit is used to perform summary calculation based on the obtained time deviation coefficient SJ, traffic congestion coefficient JT, and path transportation reliability score YS to obtain the path abnormality score PF of each traffic path, wherein the path abnormality score PF is obtained in the following manner: PF=ω1·SJ+ω2·JT+ω3·YS+C; Where ω1, ω2 and ω3 represent the weight coefficients of the time deviation coefficient SJ, the traffic congestion coefficient JT and the path transportation reliability score YS, respectively, and C represents the first correction constant; The path anomaly score threshold PFYZ is preset, and the path anomaly score PF and the path anomaly score threshold PFYZ are compared and analyzed to evaluate the path anomaly status. The specific evaluation contents are as follows: If the path anomaly score PF is greater than or equal to the path anomaly score threshold PFYZ, the transportation path is judged to be in an abnormal state. At this time, the transportation plan of this section of the path is cancelled, the path generation unit is re-performed, and the transportation path is modified; If the path anomaly score PF is less than the path anomaly score threshold PFYZ, the transportation path is judged to be in a normal state and no intervention is required.

9. A cross-border e-commerce order intelligent processing system based on blockchain according to claim 8, characterized in that: The data storage module is used to trigger the smart contract of the blockchain system after the goods arrive at the destination, automatically update the order status, complete the delivery, and store the order data, transportation route and exception handling information on the blockchain. After the order is completed, the product inventory data is updated.

10. A method for intelligently processing cross-border e-commerce orders based on blockchain, used to implement a system for intelligently processing cross-border e-commerce orders based on blockchain as claimed in any one of claims 1 to 9 above, characterized in that: The following steps are included: Step 1: Obtain the content of the consumer's order based on the e-commerce platform, and use natural language processing technology to obtain the order information, and perform structured processing on the order information to obtain a structured form of the order information, and upload the structured form of the order information to the blockchain system; Step 2: Based on the blockchain system, obtain the real-time inventory of each warehouse, and intelligently match the product delivery warehouse. After matching the delivery warehouse, use the smart contract to lock the delivery warehouse; Step 3: Build a global transportation network model and use the A algorithm to perform global path search to obtain the commodity transportation path; Step 4: Perform an abnormality assessment on each segment of the commodity transportation route. If the path assessment result is abnormal, cancel the transportation calculation of this segment of the path, re-plan the path, and optimize the commodity transportation; Step 5. After the goods are transported, the smart contract is automatically triggered to collect data related to the goods transportation and store it in the blockchain system.

Images