Cross-border logistics dynamic path planning method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of cross-border logistics dynamic path planning, and discloses a cross-border logistics dynamic path planning method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining order information of a specified cross-border order; wherein the order information at least comprises destination information and departure information of an order; generating a plurality of transportation paths based on the destination information and the departure place information; acquiring real-time data of each transportation path; inputting each transportation path and the corresponding real-time data into a preset path score model to obtain a score of each transportation path; and selecting the transportation path with the maximum score as a target transportation path according to the score of each transportation path. The method has the beneficial effects that the dynamic evaluation of the transportation path is realized, the path selection is optimized, and the safety and on-time arrival of goods are ensured, so that the overall operation efficiency of cross-border logistics is improved, and the operation cost of an enterprise is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cross-border logistics dynamic path planning, and particularly relates to a cross-border logistics dynamic path planning method and device, an electronic device and a storage medium. BACKGROUND

[0002] With the continuous development of the global economy, cross-border transportation plays an increasingly important role in international trade. However, the existing cross-border transportation path planning method mostly relies on static data, which fails to effectively combine real-time data, limiting the flexibility and accuracy of path planning. Traditional path planning is usually based on historical data, geographic information and transportation cost, etc., which cannot timely reflect the impact of climate change on transportation efficiency. For example, adverse weather conditions (such as heavy rain, storm, snow disaster, etc.) may cause transportation delay, even damage goods, bringing economic losses to enterprises. Therefore, the lack of application of real-time data will lead to the inability of further optimization of transportation route planning, resulting in reduced transportation efficiency and increased cost. SUMMARY

[0003] Therefore, it is necessary to propose a cross-border logistics dynamic path planning method, device, electronic device and storage medium for the existing cross-border logistics dynamic path planning problem.

[0004] A cross-border logistics dynamic path planning method, the method comprising: obtaining order information of a specified cross-border order; the order information at least including destination information and departure location information of the order; generating multiple transportation paths based on the destination information and the departure location information; obtaining real-time data of each transportation path; wherein the real-time data at least includes meteorological data; inputting each of the transportation paths and the corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths; selecting the transportation path with the largest score as the target transportation path according to the score of each of the transportation paths.

[0005] Further, the step of generating multiple transportation paths based on the destination information and the departure location information comprises: inputting the destination information and the departure location information into a preset mathematical model to obtain a temporary mathematical model; solving the temporary mathematical model by a preset multi-objective optimization algorithm to obtain multiple transportation paths.

[0006] Further, before the step of inputting the destination information and the departure location information into a preset mathematical model to obtain a temporary mathematical model, the method further comprises: Obtaining customs announcements of multiple countries; Analyzing the customs announcements of each country through a preset legal semantic model to obtain analyzed customs data; Inputting each of the customs data into a preset theoretical model to obtain the preset mathematical model.

[0007] Further, before the step of inputting the destination information and the departure information into a preset mathematical model to obtain a temporary mathematical model, the method further comprises: Obtaining a current date; Inputting the current date into a preset sea freight price fluctuation prediction model to obtain a Baltic Dry Index; Inputting the Baltic Dry Index and the current date into a preset theoretical model to obtain the preset mathematical model.

[0008] Further, before the step of inputting each of the transportation paths and the corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths, the method further comprises: Determining whether there is abnormal data in the real-time data that exceeds a threshold value; If there is abnormal data that exceeds the threshold value, adjusting a corresponding weight value in an initial path score model based on the abnormal data that exceeds the threshold value to obtain the preset path score model.

[0009] Further, after the step of selecting a transportation path with the largest score as a target transportation path according to the score of each of the transportation paths, the method further comprises: Monitoring policy information related to the target transportation path; Analyzing a transportation crisis index corresponding to the policy information through a preset analysis method; Determining whether the transportation crisis index is greater than a preset crisis index; If the transportation crisis index is greater than the preset crisis index, reconstructing the target transportation path.

[0010] Further, the step of obtaining real-time data of each of the transportation paths comprises: Obtaining a weather satellite cloud image; Implementing semantic segmentation on the weather satellite cloud image to extract typhoon eye radius information and moving speed information, thereby obtaining weather data.

[0011] The application further provides a cross-border logistics dynamic path planning device, which comprises: A first obtaining module configured to obtain order information of a specified cross-border order; the order information at least comprises destination information and departure information of the order; generating a plurality of transportation paths based on the destination information and the departure information; a second obtaining module, configured to obtain real-time data of each transportation path, wherein the real-time data at least includes meteorological data; an input module, configured to input each of the transportation paths and corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths; a selection module, configured to select a transportation path with the largest score as a target transportation path according to the score of each of the transportation paths.

[0012] An electronic device includes a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the following steps: obtain order information of a specified cross-border order; the order information at least includes destination information and departure information of the order; generate a plurality of transportation paths based on the destination information and the departure information; obtain real-time data of each transportation path; wherein the real-time data at least includes meteorological data; input each of the transportation paths and corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths; select a transportation path with the largest score as a target transportation path according to the score of each of the transportation paths.

[0013] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to make the processor execute the following steps: obtain order information of a specified cross-border order; the order information at least includes destination information and departure information of the order; generate a plurality of transportation paths based on the destination information and the departure information; obtain real-time data of each transportation path; wherein the real-time data at least includes meteorological data; input each of the transportation paths and corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths; select a transportation path with the largest score as a target transportation path according to the score of each of the transportation paths.

[0014] The beneficial effects of the present application: by obtaining and analyzing the destination and departure information of specific orders, generating multiple transportation paths, and monitoring the weather conditions in real time, inputting these data into the preset path score model, thereby realizing the dynamic evaluation of the transportation path, optimizing the path selection, ensuring the safety and timely arrival of the goods, thereby improving the overall operation efficiency of cross-border logistics, reducing the operating cost of enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0016] Among them: Figure 1 It is an application environment diagram of the cross-border logistics dynamic path planning method in an embodiment; Figure 2 It is a flowchart of the cross-border logistics dynamic path planning method in an embodiment; Figure 3 It is a structural block diagram of the cross-border logistics dynamic path planning device in an embodiment; Figure 4 It is a structural block diagram of the electronic device in an embodiment. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0018] Figure 1 It is an application environment diagram of the cross-border logistics dynamic path planning in an embodiment. Referring to Figure 1 The cross-border logistics dynamic path planning method is applied to a cross-border logistics dynamic path planning system. The cross-border logistics dynamic path planning system includes a terminal 110 and a server 120. The terminal 110 and the server 120 are connected through a network, and the terminal 110 can be a desktop terminal or a mobile terminal, and the mobile terminal can be at least one of a mobile phone, a tablet computer, a notebook computer, etc. The server 120 can be realized by an independent server or a server cluster composed of multiple servers. The terminal 110 is used to obtain real-time data, and the server 120 is used to generate target transportation paths.

[0019] AsFigure 2 As shown in one embodiment, a cross-border logistics dynamic path planning method is provided. The method can be applied to terminals and servers, and the embodiment is exemplified by application to terminals. The cross-border logistics dynamic path planning method specifically includes the following steps: S1: obtaining order information of a specified cross-border order; the order information at least includes destination information and departure location information of the order; S2: generating multiple transportation paths based on the destination information and the departure location information; S3: obtaining real-time data of each transportation path; wherein the real-time data at least includes meteorological data; S4: inputting each transportation path and the corresponding real-time data into a preset path score model to obtain a score of each transportation path; S5: selecting a transportation path with the largest score as a target transportation path according to the score of each transportation path.

[0020] As described in step S1 above, the order information of a specified cross-border order is obtained; the order information at least includes destination information and departure location information of the order. The order information can also specifically include the type, quantity and size of the goods, and the destination information refers to the final delivery location of the goods, including country, city, postal code, etc. The departure location information refers to the starting location of the goods, also containing relevant information such as country and city, clearly indicating the starting and ending points of transportation, providing necessary geographic basic data for generating transportation paths. In addition, other relevant information of the order can also be considered, such as the type, weight, volume and value of the goods, or even the special requirements or time limit of the customer, in order to better meet the customer's needs and improve the overall logistics operation efficiency.

[0021] As described in step S2 above, multiple transportation paths are generated based on the destination information and the departure location information. According to the information of the departure location and the destination, map data and traffic network are used to generate multiple feasible transportation paths. The generation process needs to consider various factors, such as the availability and efficiency of different transportation modes such as land, sea and air transportation, covering multiple alternative routes, and through algorithm calculation (such as the shortest path algorithm, Dijkstra algorithm or A* algorithm, etc.), different transportation channels, passing cities, traffic hubs, etc. can be identified, and factors such as traffic conditions, border clearance time and possible operation limits of logistics companies need to be considered. Thus, diversified path selection is provided, giving decision makers more options, so as to select the optimal path in subsequent analysis, ensuring that the goods arrive at the destination on time and safely.

[0022] As described in step S3 above, real-time data of each transportation path is obtained. After generating the possible transportation paths, real-time data related to each path is obtained. These real-time data mainly include meteorological data such as temperature, precipitation, wind speed, haze and storm warning information, and can also include real-time customs policies, freight data, port container yard occupancy rate, and real-time data acquisition depends on meteorological bureau, satellite remote sensing, weather forecast API and other data sources. Real-time meteorological data can help the system to judge the transportation safety and efficiency of each path, for example, in bad weather conditions, some transportation paths may face delays or safety risks, and even need to be temporarily closed, therefore, obtaining real-time data not only helps to identify potential risks, but also provides the basis for decision-making, so that the subsequent path scoring is more accurate, greatly improving the scientificity and effectiveness of the overall path planning.

[0023] As described in step S4 above, each transportation path and the corresponding real-time data are input into a preset path score model to obtain the score of each transportation path. After collecting each transportation path and its corresponding real-time data, these information is input into a preset path score model. The construction of the model is based on machine learning or expert system, which evaluates various attributes of transportation paths through historical data analysis and pattern recognition. The scoring model will consider multiple factors, including path distance, transportation time, real-time weather influence, traffic conditions, transportation cost, etc. Each path will be given a score according to these indicators, and the higher the score, the more obvious the comprehensive advantages of this path under the current weather conditions and traffic conditions, so that the scoring standard can be dynamically adjusted to adapt to different types of transportation demand and external environmental changes, ensuring that the finally selected path can achieve the best effect in actual operation. It should be noted that the preset path score model can be obtained by training a first neural network model based on a preset sample set. Each sample data in the preset sample set includes sample path data (including transportation path and corresponding real-time data) and path score corresponding to the sample path data. When training the pre-constructed first neural network model, the sample path data in each sample data is used as the input of the first neural network model, and the path score corresponding to the sample path data in each sample data is used as the output of the first neural network model. Through training, the first neural network model can learn the corresponding relationship between all possible sample path data and path score. The trained first neural network model is used as the preset path score model.

[0024] As described in step S5, the transport path with the highest score is selected as the target transport path according to the score of each transport path. By analyzing the preset path score model, the score of each transport path can be obtained. At this time, according to the score of each path, the path with the highest score is selected as the target transport path, thereby realizing automatic decision-making. In a specific embodiment, the path with the highest score can also be adjusted in combination with the specific needs of the user and the actual transport situation. The path with the highest score is usually the optimal choice after considering various factors. If the user may have other considerations, such as cost control and timeliness requirements. Therefore, even if the system recommends the optimal path, the final decision may need to be confirmed or adjusted manually, ensuring the organic combination of intelligent decision-making and human experience, making the path planning of cross-border transportation more scientific and efficient.

[0025] In an embodiment, the step S2 of generating a plurality of transport paths based on the destination information and the departure information comprises: S201: inputting the destination information and the departure information into a preset mathematical model to obtain a temporary mathematical model; S202: solving the temporary mathematical model by a preset multi-objective optimization algorithm to obtain a plurality of transport paths.

[0026] As described in steps S201-S202, the destination information and the departure information are input into a preset mathematical model to obtain a temporary mathematical model. The destination information and the departure information obtained from the cross-border order can be converted into a data format that can be processed mathematically before input. The preset mathematical model usually includes an objective function, a constraint condition and a variable, which can reflect the connection relationship between the departure and the destination, the possible transport mode and the related cost factors. Specifically, the destination and departure information are input by coordinate (such as latitude and longitude), forming corresponding nodes. Between these nodes, the model will set edges to represent possible transport paths and assign each edge a corresponding cost, such as distance, time or cost, etc. These costs can be represented by linear or nonlinear functions to adapt to different transport situations. Finally, the mathematical model will generate a temporary model that can cover multiple paths from the starting point to the end point for subsequent optimization algorithms.

[0027] The temporary mathematical model is solved by a preset multi-objective optimization algorithm to obtain multiple transportation paths. The previously generated temporary mathematical model is solved by a preset multi-objective optimization algorithm to obtain multiple transportation paths. The multi-objective optimization algorithm is a kind of algorithm for processing multiple optimization objectives, which can consider multiple factors in an optimization process to find multiple optimal solutions, and is suitable for processing complex problems such as diversified transportation path requirements. In actual application, the preset multi-objective optimization algorithm can be any one of genetic algorithm, ant colony algorithm or particle swarm optimization algorithm. These algorithms can effectively explore the solution space and gradually approach the optimal solution through operations such as crossover, selection and mutation. In the solving process, the algorithm evaluates the comprehensive performance of each path, and the factors considered can include transportation time, cost, reliability, risk, etc. Finally, the multiple transportation paths output by the algorithm will be sorted according to the set objective function to select the optimal path combination. This diversified path selection enables decision makers to flexibly adjust according to actual requirements, improves transportation efficiency, reduces potential risks, and ensures optimal decision-making in a dynamic environment. It should be noted that in the preset mathematical model, various data are pre-stored, such as customs policies of various countries, oil prices, Baltic indices, etc. Of course, it can also include some real-time data, such as real-time policies, emergencies, weather data, etc. Since various different real-time data have been considered in the aforementioned generation of multiple paths, real-time data need not be considered here.

[0028] In one embodiment, before the step S201 of inputting the destination information and the departure information into a preset mathematical model to obtain a temporary mathematical model, the method further comprises: S2001: obtaining customs announcements of multiple countries; S2002: analyzing the customs announcements of each country by a preset legal semantic model to obtain analyzed customs data; S2003: inputting each of the customs data into a preset theoretical model to obtain the preset mathematical model.

[0029] As described in the above steps S2001-S2003, collect customs announcements of multiple countries related to cross-border transportation. These announcements usually contain key information such as laws and regulations on import and export of goods, customs policies, inspection and quarantine requirements, embargo and restricted commodity lists, etc. In order to ensure the comprehensiveness and timeliness of the data, the process of obtaining these customs announcements may involve multiple ways through official channels (such as government customs websites), specialized logistics service platforms, legal and regulatory databases, etc., to ensure that current policy changes and implementation details can be covered. By comprehensively understanding the customs policies of various countries, the compliance, economy and potential risks of the transportation path can be more effectively evaluated, providing early warning for compliance issues that may be encountered during transportation, helping logistics decision-makers better avoid risks and reduce potential economic losses when developing transportation plans.

[0030] The collected customs announcements are parsed using a preset legal semantic model. The legal semantic model is a natural language processing (NLP) based algorithm tool that can understand and analyze the specialized terms, structure and grammar in legal texts, extracting useful information. In the process of parsing customs announcements, this model will identify key clauses, policies and data information such as customs rates, import restrictions, reporting requirements, etc. The parsing process includes text preprocessing, word segmentation, syntax analysis, information extraction and other steps to ensure that important information in the announcement can be accurately captured. The parsed customs data will be stored in a structured form for subsequent analysis and processing. Ensuring that subsequent path planning decisions are based on compliance can help avoid legal risks or increased operating costs due to policy changes during the transportation of goods.

[0031] The parsed customs data is input into a preset theoretical model to construct a comprehensive preset mathematical model. This mathematical model not only reflects the origin and destination information, but also integrates policy information from various national customs. This is the core of path planning, because the choice of path is not only affected by geographical factors, but also needs to consider the compliance of customs policies. The theoretical model includes objective functions and constraints, which is based on existing theoretical framework and mathematical derivation, used to describe the basic principles of path planning, relying on pre-set planning logic, these elements will consider the customs announced tariffs, trade restrictions, legal procedures and other compliance requirements related to cross-border transportation, by integrating customs data into the model, the logistics system can more accurately simulate the actual transportation situation, and fully evaluate the performance of different path options. In this way, modeling can provide more scientific decision-making basis for subsequent multi-objective optimization algorithms. Finally, the obtained preset mathematical model not only provides a structured framework for subsequent transportation path generation, but also ensures that the designed path scheme meets the policy requirements of multiple national customs, while optimizing cost and time efficiency. This process not only improves the accuracy of path planning, but also provides solid data support for actual business response.

[0032] In one embodiment, before the step S201 of inputting the destination information and the origin information into a preset mathematical model to obtain a temporary mathematical model, the method further comprises: S2101: obtaining a current date; S2102: inputting the current date into a preset maritime price fluctuation prediction model to obtain a Baltic Dry Index; S2103: inputting the Baltic Dry Index and the current date into a preset theoretical model to obtain the preset mathematical model.

[0033] As described in steps S2101-S2103 above, the current date information is obtained. The current date is a key information in cross-border transportation and logistics management, which affects the cost of goods transportation, time planning and market price changes, etc. The date can be obtained through the system's clock function to ensure real-time and accuracy. In the field of transportation price fluctuation and customs policy, timeliness is particularly important. For example, transportation prices will fluctuate due to season, holidays, market demand, etc. Customs management policies and regulations may also be adjusted over time. Therefore, this date information will become an important parameter input into the price fluctuation prediction model, ensuring that subsequent decisions are based on the latest market situation, helping managers make more reasonable judgments in a dynamic environment.

[0034] The current date is input into a preset sea freight price fluctuation prediction model. This prediction model is designed to use historical data and market dynamic information to model and predict sea freight price trends. Specifically, this model usually considers multiple influencing factors, including supply and demand relationship, seasonal changes, international political and economic environment, natural disasters, etc., to help predict future BDI trends. By inputting the current date, the model can identify recent historical data and make predictions based on this data. The Baltic Dry Index is a very important indicator in the international shipping market, reflecting the price level of global dry bulk shipping, mainly used to measure the market price of dry bulk. Changes in this index can greatly affect logistics costs, thereby affecting the selection of transportation routes and cost budgeting. Therefore, obtaining and analyzing this index not only helps logistics companies effectively manage their budgets, but also provides important economic basis for optimizing route selection and supporting the rationality and efficiency of logistics decisions.

[0035] The predicted BDI and the current date are input into a preset theoretical model, which includes an objective function and various constraints such as transportation time, cost, compliance, etc. When integrating the BDI, the index will be used as an important parameter in the model to reflect the economic efficiency of the transportation route. By introducing this market price index into the model, the system can more accurately assess transportation costs under different routes, providing data support for subsequent route planning, and the resulting preset mathematical model will be able to more comprehensively and accurately reflect the transportation routes between the origin and destination and their economic benefits, providing ideal data support for subsequent route generation and optimization.

[0036] In a specific embodiment, the BDI obtained above, the current date, and customs data can also be input into a preset theoretical model to obtain a more comprehensive mathematical model that can provide better data support for transportation route generation.

[0037] In one embodiment, before the step S4 of inputting each of the transportation routes and the corresponding real-time data into a preset path score model to obtain a score for each of the transportation routes, the method further comprises: S301: Determine whether there is abnormal data in the real-time data that exceeds a threshold value; S302: If there is abnormal data that exceeds the threshold value, adjust the corresponding weight value in the initial path score model based on the abnormal data that exceeds the threshold value to obtain the preset path score model.

[0038] As described in steps S301-S302, the acquired real-time data is evaluated to determine whether there are abnormal data. When the real-time data contains abnormal values, it may lead to inaccurate results of the preset path scoring model, thereby affecting the selection of the transportation path and further affecting the efficiency and cost of logistics. Abnormal data can include extreme values in meteorological data such as temperature changes, wind speed, and precipitation, or other indicators that may affect transportation safety and efficiency, such as typhoon grade greater than 12, tax rate mutation greater than 5%, and fuel price increase greater than 8%. Therefore, the system needs to set reasonable thresholds to quickly identify these abnormal data. The setting of thresholds is usually based on historical data and actual experience of the business to ensure that it has certain pertinence and adaptability. Based on the detected abnormal data, the initial path scoring model is adjusted. The specific process includes analyzing the abnormal data and adjusting the weights of relevant factors in the scoring model according to their nature and possible impact. During the adjustment process, the abnormal data is considered as feedback to the model to correct the model's response in specific situations. For example, if a certain transportation path shows high risk under certain weather conditions (such as heavy rain or strong wind), a dynamic factor matrix is set, and weight adjustment is achieved through the dynamic factor matrix. The matrix contains three types of modalities: weather, policy, and cost. Each modality type has a basic weight, which can be dynamically adjusted according to the corresponding abnormal data. Specifically, when the typhoon grade is greater than or equal to 12, the corresponding weight is increased to 65%; when the tax rate mutation is greater than 5% (±5%), the weight is increased to 50%; and when the fuel price increases by more than 8% in a single day, the weight is reduced to 30%. Thus, dynamic optimization is achieved to ensure that the model still has good adaptability when facing uncertainty.

[0039] In one embodiment, after the step S5 of selecting the transportation path with the highest score as the target transportation path, the method further comprises: S601: monitoring policy information related to the target transportation path; S602: analyzing the transportation crisis index corresponding to the policy information by a preset analysis method; S603: determining whether the transportation crisis index is greater than a preset crisis index; S604: if the transportation crisis index is greater than the preset crisis index, reconstructing the target transportation path.

[0040] As mentioned in steps S601-S604, the policy information related to the target transportation path is continuously monitored. Ensuring the smooth progress of the entire transportation chain, as cross-border transportation often involves multiple national laws, regulations, and policy requirements, these factors can have a significant impact on the transportation plan. Policy information may include customs policies, import / export restrictions, changes in tariffs, transportation regulations, and other regulations related to the path. The process of monitoring policy information can be achieved through automated data scraping tools or API interfaces, regularly updated to obtain the latest information. This information can come from government websites, industry associations, logistics management platforms, and other sources. By establishing a dynamic monitoring system, logistics managers can keep abreast of the latest policy changes, thereby taking appropriate measures to ensure compliance and efficiency during transportation. The importance of this step lies in its ability to identify potential policy risks in advance, helping enterprises adjust transportation strategies in a timely manner to avoid potential legal risks and cost losses, thereby ensuring the smooth progress of transportation. In-depth analysis of the monitored policy information to calculate the transportation crisis index related to the target transportation path can use pre-set analysis methods, such as quantitative analysis, situational analysis, or risk assessment models, thereby converting complex policy information into a numerical crisis index for subsequent decision-making. The transportation crisis index is a comprehensive indicator designed to assess the risk level that a particular transportation path may face. It usually considers multiple factors, including the stability of relevant policies, economic environment, market supply and demand conditions, and other relevant external risk variables. This indicator not only reflects the direct impact of policies, but may also cover indirect impacts, such as sudden increases in transportation costs or potential delays. By analyzing the relationship between policy information and the transportation crisis index, the system can provide a basis for decision-making, allowing decision-makers to adjust transportation plans in a timely manner to address potential crises. This analysis process not only improves the scientific nature of monitoring, but also provides a quantitative tool for subsequent risk management, thereby playing a crucial role in information-based decision-making. Determine whether the calculated transportation crisis index exceeds a pre-set crisis index. This pre-set index is usually based on historical data, industry standards, the enterprise's own risk tolerance, and specific transportation conditions, and is a key reference point for assessing the feasibility and safety of the transportation path. If the transportation crisis index exceeds the pre-set crisis index, the system will reconstruct the target transportation path. By assessing the potential risks caused by factors such as policy changes, the current transportation plan is redesigned to ensure safe and efficient completion of the transportation task in the new policy environment. Reconstructing the target transportation path may include various strategies, such as selecting alternative transportation routes, adjusting transportation methods (such as changing from sea transport to air transport), optimizing loading plans, or even changing suppliers or destinations. This reconstruction process needs to combine real-time data, policy analysis, and market conditions to ensure that the new path chosen not only reduces risks, but also remains competitive in terms of cost and time.In one embodiment, a multimodal transport strategy tree construction can be established based on a target transport path, taking the target transport path as the main layer, obtaining alternative schemes of the target transport path in various regions as the branch layer. The alternative schemes can be generated in the manner of generating path paths with alternative options according to the main cities or nodes passed through by the target transport path. For example, when the target transport path passes through a certain main city, different transport modes and routes from the city to adjacent cities can be considered. Specifically, all key nodes (such as ports, railway stations, aviation hubs, etc.) in the target path are identified, and for each key node, multiple paths from the city to adjacent cities are analyzed, which can involve highways, railways or waterways, and historical transport data is used to analyze which paths perform better under different conditions to generate alternative schemes. In addition, geopolitical mutation response plans can be set based on the country or city corresponding to the target transport path as the emergency layer.

[0041] In one embodiment, the step S3 of obtaining real-time data of each transport path comprises: S311: Obtain a weather satellite cloud image; S312: Perform semantic segmentation on the weather satellite cloud image to extract typhoon eye radius information and moving speed information, thereby obtaining meteorological data.

[0042] As described in steps S311-S312 above, it is necessary to obtain the latest satellite cloud images from meteorological satellites. Meteorological satellite images are images of the Earth's surface and atmospheric cloud layers taken by meteorological satellites, which contain rich meteorological information such as temperature, humidity, wind speed, air pressure, and the distribution of weather systems. Select a reliable meteorological satellite data source, such as the satellite data provided by the Meteorological Bureau, NASA, NOAA, and other public meteorological data. These data sources usually have high timeliness and accuracy. Data access: Use API interface or directly download the latest satellite cloud images from the meteorological agency's database. Modern meteorological data platforms usually provide interfaces to support automated data acquisition. The obtained meteorological satellite cloud images will be subjected to semantic segmentation processing to extract specific meteorological data, such as the radius information and movement speed information of the typhoon eye. Semantic segmentation is a computer vision technology, the main purpose of which is to label each pixel in the image as a specific category, thereby realizing the identification and analysis of regions. Before semantic segmentation, the satellite cloud image needs to be processed first, including image scaling, denoising, contrast adjustment, etc., to improve the effect and accuracy of subsequent segmentation. Use a deep learning model (such as U-Net, SegNet, etc.) for training. The model needs to be trained on cloud images with labeled data to learn how to segment specific phenomena (such as typhoons, rainstorm clouds, etc.) from images. The processed satellite cloud image is predicted using the trained model to label the typhoon eye and related meteorological feature areas. The model classifies each pixel point to determine the typhoon eye, cold cloud area, warm cloud area, and other different areas. Data extraction: Extract specific data such as the radius and movement speed information of the typhoon eye from the semantic segmentation results, which will be integrated into a structured data format for subsequent use. In transportation management, extreme weather events such as typhoons can have a significant impact on transportation safety, so timely extraction and analysis of this information will help logistics companies and decision-makers develop safer and more reasonable response strategies.

[0043] Referring to Figure 3 The application also provides a cross-border logistics dynamic path planning device, which comprises: A first acquisition module 902 is configured to acquire order information of a specified cross-border order, wherein the order information at least includes destination information and departure location information of the order. A generation module 904 is configured to generate a plurality of transportation paths based on the destination information and the departure location information. A second acquisition module 906 is configured to acquire real-time data of each transportation path, wherein the real-time data at least includes meteorological data. An input module 908 is configured to input each transportation path and corresponding real-time data into a preset path score model to obtain a score of each transportation path. The selecting module 910 is configured to select a transportation path with the largest score as the target transportation path according to the score of each transportation path.

[0044] In an embodiment, the generating module 904 includes: An information input sub-module is configured to input the destination information and the departure information into a preset mathematical model to obtain a temporary mathematical model. A solving sub-module is configured to solve the temporary mathematical model by using a preset multi-objective optimization algorithm to obtain a plurality of transportation paths.

[0045] In an embodiment, the generating module 904 further includes: A customs announcement obtaining sub-module is configured to obtain customs announcements of a plurality of countries. A customs announcement analyzing sub-module is configured to analyze the customs announcements of the plurality of countries by using a preset legal semantic model to obtain analyzed customs data. A customs data input sub-module is configured to input the customs data into a preset theoretical model to obtain the preset mathematical model.

[0046] In an embodiment, the generating module 904 further includes: A current date obtaining sub-module is configured to obtain a current date. A current date first input sub-module is configured to input the current date into a preset sea freight price fluctuation prediction model to obtain a Baltic Dry Index. A current date second input sub-module is configured to input the Baltic Dry Index and the current date into a preset theoretical model to obtain the preset mathematical model.

[0047] In an embodiment, the cross-border logistics dynamic path planning apparatus further includes: An abnormal data judging module is configured to judge whether there is abnormal data in the real-time data that exceeds a threshold. A weight value adjusting module is configured to, if there is abnormal data that exceeds the threshold, adjust a corresponding weight value in an initial path score model based on the abnormal data that exceeds the threshold to obtain the preset path score model.

[0048] In an embodiment, the cross-border logistics dynamic path planning apparatus further includes: A policy information monitoring module is configured to monitor policy information related to the target transportation path. A transportation crisis index analyzing module is configured to analyze a transportation crisis index corresponding to the policy information by using a preset analysis method. A transportation crisis index judging module is configured to judge whether the transportation crisis index is greater than a preset crisis index. The target transportation route reconstruction module is used to reconstruct the target transportation route if the risk level is greater than the preset risk index.

[0049] In one embodiment, the second acquisition module 906 includes: The meteorological satellite cloud image acquisition submodule is used to acquire meteorological satellite cloud images; The meteorological satellite cloud image semantic segmentation submodule is used to perform semantic segmentation on the meteorological satellite cloud image to extract typhoon eye radius information and movement speed information, thereby obtaining meteorological data.

[0050] Figure 4 An internal structural diagram of an electronic device in one embodiment is shown. This electronic device can specifically be a terminal or a server, and more specifically, a computer device. Figure 4 As shown, the electronic device includes a processor, a memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement a cross-border logistics dynamic path planning method. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to implement the cross-border logistics dynamic path planning method. Those skilled in the art will understand that… Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0051] In one embodiment, an electronic device is provided, including a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps: Obtain the order information for a specified cross-border order; the order information includes at least the destination information and the origin information of the order; Multiple transportation routes are generated based on the destination information and the origin information; Acquire real-time data for each transportation route; wherein the real-time data includes at least meteorological data; Each of the aforementioned transportation routes and its corresponding real-time data are input into a preset route scoring model to obtain the score for each of the aforementioned transportation routes; Based on the scores of each transport route, the transport route with the highest score is selected as the target transport route.

[0052] By acquiring and analyzing the destination and departure information of specific orders, multiple transportation paths are generated, and meteorological conditions are monitored in real time. These data are input into a preset path score model to realize dynamic evaluation of the transportation path, optimize path selection, ensure the safety and timely arrival of goods, thereby improving the overall operation efficiency of cross-border logistics and reducing enterprise operation costs.

[0053] In one embodiment, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the processor performs the following steps: acquiring order information of a specified cross-border order; the order information at least includes destination information and departure information of the order; generating multiple transportation paths based on the destination information and the departure information; acquiring real-time data of each transportation path; wherein the real-time data at least includes meteorological data; inputting each of the transportation paths and the corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths; selecting the transportation path with the largest score as the target transportation path according to the score of each of the transportation paths.

[0054] By acquiring and analyzing the destination and departure information of specific orders, multiple transportation paths are generated, and meteorological conditions are monitored in real time. These data are input into a preset path score model to realize dynamic evaluation of the transportation path, optimize path selection, ensure the safety and timely arrival of goods, thereby improving the overall operation efficiency of cross-border logistics and reducing enterprise operation costs.

[0055] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiment methods can be included. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0056] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0057] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A dynamic path planning method for cross-border logistics, characterized in that, The method comprises: obtaining order information of a specified cross-border order; the order information at least comprises destination information and departure location information of the order; generating a plurality of transportation paths based on the destination information and the departure location information; obtaining real-time data of each transportation path; wherein the real-time data at least comprises meteorological data; inputting each of the transportation paths and the corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths; selecting a transportation path with the largest score as a target transportation path according to the score of each of the transportation paths.

2. The cross-border logistics dynamic path planning method according to claim 1, wherein, The step of generating a plurality of transportation paths based on the destination information and the departure location information comprises: inputting the destination information and the departure location information into a preset mathematical model to obtain a temporary mathematical model; solving the temporary mathematical model by a preset multi-objective optimization algorithm to obtain a plurality of transportation paths. 3.The cross-border logistics dynamic path planning method of claim 2, wherein, Before the step of inputting the destination information and the departure location information into a preset mathematical model to obtain a temporary mathematical model, the method further comprises: obtaining customs announcements of a plurality of countries; analyzing the customs announcements of the countries by a preset legal semantic model to obtain analyzed customs data; inputting each of the customs data into a preset theoretical model to obtain the preset mathematical model.

4. The cross-border logistics dynamic path planning method according to claim 2, wherein, Before the step of inputting the destination information and the departure location information into a preset mathematical model to obtain a temporary mathematical model, the method further comprises: obtaining a current date; inputting the current date into a preset sea freight price fluctuation prediction model to obtain a Baltic Dry Index; inputting the Baltic Dry Index and the current date into a preset theoretical model to obtain the preset mathematical model.

5. The cross-border logistics dynamic path planning method according to claim 1, wherein, Before the step of inputting each of the transportation paths and the corresponding real-time data into a preset path score model to obtain a score of each of the transportation paths, the method further comprises: determining whether there is abnormal data beyond a threshold value in the real-time data; if there is abnormal data beyond the threshold value, adjusting a corresponding weight value in an initial path score model based on the abnormal data beyond the threshold value to obtain the preset path score model.

6. The cross-border logistics dynamic path planning method according to claim 1, wherein, After the step of selecting a transportation path with the largest score as a target transportation path according to the score of each of the transportation paths, the method further comprises: monitoring policy information related to the target transportation path; analyzing a transportation crisis index corresponding to the policy information by a preset analysis method; determining whether the transportation crisis index is greater than a preset crisis index; if the transportation crisis index is greater than the preset crisis index, reconstructing the target transportation path.

7. The cross-border logistics dynamic path planning method according to claim 1, wherein, The step of obtaining real-time data of each transportation path comprises: obtaining a meteorological satellite cloud image; implementing semantic segmentation on the meteorological satellite cloud image to extract typhoon eye radius information and moving speed information to obtain meteorological data.

8. A dynamic path planning device for cross-border logistics, characterized in that, The device comprises: a first obtaining module configured to obtain order information of a specified cross-border order; the order information at least comprises destination information and departure location information of the order; a generating module configured to generate a plurality of transportation paths based on the destination information and the departure location information; a first obtaining module configured to obtain order information of a specified cross-border order; the order information at least comprises destination information and departure location information of the order; A second obtaining module is configured to obtain real-time data of each transportation path, wherein the real-time data at least includes meteorological data; An input module is configured to input each transportation path and corresponding real-time data into a preset path score model to obtain a score of each transportation path; A selection module is configured to select a transportation path with the largest score as a target transportation path according to the score of each transportation path.

9. A computer-readable storage medium, characterized in that, A computer program is stored, and the computer program is executed by a processor to make the processor execute the steps of the cross-border logistics dynamic path planning method according to any one of claims 1 to 7.

10. An electronic device, comprising: The device includes a memory and a processor, and the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the cross-border logistics dynamic path planning method according to any one of claims 1 to 7.

Images