Cross-border logistics demand prediction method and system

Abstract

The invention relates to the technical field of logistics management, in particular to a cross-border logistics demand prediction method and system, which can provide accurate and reliable prediction results and decision support for cross-border logistics enterprises. The method comprises the following steps: acquiring a destination cargo demand influence parameter set and a delivery place cargo supply influence parameter set; respectively performing feature analysis on the destination cargo demand influence parameter set and the shipping place cargo supply influence parameter set to obtain a destination cargo demand feature index and a shipping place cargo supply feature index; comparing the destination cargo demand characteristic index with the shipping place cargo supply characteristic index, extracting a minimum characteristic index, and marking the minimum characteristic index as a logistics demand characteristic index; obtaining the real-time cross-border transportation cost and the historical same-period cargo transportation characteristic quantity corresponding to the logistics demand characteristic index; and inputting the logistics demand characteristic index and the real-time cross-border transportation cost into a pre-constructed cross-border logistics demand influence analysis model as input.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a cross-border logistics demand forecasting method and system. Background Art

[0002] In the context of globalization, cross-border logistics has become an important link connecting trade between different countries and regions. With the development of e-commerce, more and more companies and individuals are participating in cross-border commodity transactions, which has also led to a sharp increase in the demand for cross-border logistics.

[0003] Existing logistics demand forecasting methods are often based on single-dimensional data analysis, such as historical sales data, seasonal factors, etc., but in the context of cross-border logistics, this method is too simple. In actual operations, logistics demand is not only affected by market factors at the destination, but also by market factors at the place of shipment. For example, policy changes, seasonal demand changes, and economic conditions in the destination country will affect the demand for imported goods; similarly, policies, seasonality, and economic factors in the country of shipment will also affect the exporter's supply capacity.

[0004] At present, many cross-border logistics companies still rely on empirical methods to forecast logistics demand, which often seems powerless in the face of emergencies. For example, if the place of shipment suddenly introduces new export restrictions, or the destination country encounters an economic crisis, it will be difficult for existing forecasting methods to adjust logistics plans in a timely and accurate manner.

[0005] Therefore, there is an urgent need for a cross-border logistics demand forecasting method that can effectively integrate various influencing factors and provide enterprises with more accurate logistics demand forecasts. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a cross-border logistics demand forecasting method and system which can provide accurate and reliable forecasting results and decision support for cross-border logistics enterprises.

[0007] In a first aspect, the present invention provides a method for predicting cross-border logistics demand, the method comprising:

[0008] Collect the parameter set affecting the cargo demand at the destination and the parameter set affecting the cargo supply at the place of shipment;

[0009] Perform characteristic analysis on the parameter set influencing the cargo demand at the destination and the parameter set influencing the cargo supply at the place of shipment respectively, and obtain the characteristic index of the cargo demand at the destination and the characteristic index of the cargo supply at the place of shipment;

[0010] Compare the destination cargo demand characteristic index and the origin cargo supply characteristic index, extract the smallest characteristic index, and mark it as the logistics demand characteristic index;

[0011] Obtain the real-time cross-border transportation cost and the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index;

[0012] The logistics demand characteristic index and real-time cross-border transportation cost are used as inputs into the pre-built cross-border logistics demand impact analysis model to obtain the cross-border logistics demand factor;

[0013] Based on the cross-border logistics demand factor, the characteristic quantities of freight transportation in the same period in history are corrected to obtain the forecast quantity of cross-border logistics demand.

[0014] Furthermore, the mathematical formula for calculating the cross-border logistics demand factor by the cross-border logistics demand impact analysis model is:

[0015] F=w 1 I+w 2 ·C+b

[0016] Among them, F represents the cross-border logistics demand factor, I represents the logistics demand characteristic index, C represents the real-time cross-border transportation cost, and w 1 represents the impact of the logistics demand characteristic index on the cross-border logistics demand factor, w 2 It represents the influence of real-time cross-border transportation cost on cross-border logistics demand factor, and b represents the bias term, which is used to adjust the baseline prediction value of the model.

[0017] Furthermore, the method for constructing the cross-border logistics demand impact analysis model includes:

[0018] Collect historical data, including logistics demand characteristic index, real-time cross-border transportation costs and historical logistics demand in different time periods;

[0019] Clean the collected historical data, remove duplicate data, process missing values, and identify and process outliers;

[0020] Selecting machine learning models as the infrastructure of the cross-border logistics demand impact analysis model, the machine learning models include random forest, gradient boosting tree, support vector machine, neural network and recurrent neural network;

[0021] Divide the processed data into a training set and a validation set, and train the selected model using the training set;

[0022] Optimize the performance of the model on the validation set by adjusting the model parameters;

[0023] Use an independent test set to evaluate the model's predictive performance and adjust the model structure, parameters, and feature engineering scheme based on the evaluation results;

[0024] Deploy the trained model to production environment.

[0025] Furthermore, the method for obtaining the destination cargo demand characteristic index and the place of shipment cargo supply characteristic index includes:

[0026] Check the parameter sets affecting the demand for goods at the destination and the parameter sets affecting the supply of goods at the origin and deal with errors, duplicate records and inconsistencies in the data;

[0027] Analyze each influencing parameter separately, identify the commonalities among multiple factors, and extract key features;

[0028] Convert qualitative descriptions of policy impacts and seasonal changes into concrete numerical indicators;

[0029] Weights are assigned to each factor according to its importance and influence, and then the weighted average is calculated as the characteristic index.

[0030] Furthermore, the method for obtaining the real-time cross-border transportation cost includes:

[0031] Identify logistics providers as data sources;

[0032] Use logistics information platform services to obtain cost information;

[0033] Establish data interface with logistics information platform service;

[0034] Obtain logistics parameters through logistics information platform services and calculate real-time cross-border transportation costs based on logistics parameters;

[0035] Verify the real-time cross-border transportation cost data obtained and clean the noise and outliers in the data;

[0036] The verified real-time cross-border transportation cost data is stored in the database.

[0037] Furthermore, the method for obtaining the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index in the same period includes:

[0038] Extracting shipping data from the company’s historical transaction records;

[0039] Categorize data by date to allow for year-over-year comparisons;

[0040] Process and analyze historical data to obtain patterns and trends in historical data, i.e., freight transportation characteristics;

[0041] By comparing the logistics demand characteristic index, select the matching historical freight transportation characteristic quantities for the same period.

[0042] Furthermore, the method for obtaining the cross-border logistics demand forecast includes:

[0043] Set the correction coefficient according to the specific value of the cross-border logistics demand factor;

[0044] Select the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index as the benchmark;

[0045] Apply the correction coefficient to adjust the selected historical freight transportation characteristic quantity in the same period to obtain the corrected logistics demand forecast quantity;

[0046] Use the latest actual logistics data to verify the corrected logistics demand forecast and evaluate the accuracy of the forecast;

[0047] The verified and adjusted cross-border logistics demand forecast is used as the final output.

[0048] On the other hand, the present application also provides a cross-border logistics demand forecasting system, the system comprising:

[0049] Data collection module, collecting the parameter set affecting the cargo demand at the destination and the parameter set affecting the cargo supply at the place of shipment;

[0050] The characteristic analysis module performs characteristic analysis on the parameter set affecting the demand of the destination goods and the parameter set affecting the supply of the goods at the place of shipment, respectively, to obtain the characteristic index of the demand of the destination goods and the characteristic index of the supply of the goods at the place of shipment;

[0051] The index comparison module compares the destination cargo demand characteristic index and the origin cargo supply characteristic index, extracts the smallest characteristic index, and marks it as the logistics demand characteristic index;

[0052] The historical data association module obtains the real-time cross-border transportation cost and the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index;

[0053] The demand impact analysis module takes the logistics demand characteristic index and the real-time cross-border transportation cost as inputs into the pre-built cross-border logistics demand impact analysis model to obtain the cross-border logistics demand factor;

[0054] The demand forecast correction module corrects the historical freight transportation characteristic quantities in the same period based on the cross-border logistics demand factors to obtain the cross-border logistics demand forecast.

[0055] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.

[0056] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.

[0057] Compared with the prior art, the present invention has the following beneficial effects: the method not only takes into account traditional historical sales data and seasonal factors, but also deeply analyzes multi-dimensional influencing parameters such as policies, seasons and economy of the destination and the place of shipment, so that the forecast can more accurately reflect the changes in actual logistics demand;

[0058] By collecting and analyzing influencing parameters in real time, this method can respond quickly to market and policy changes; when the policy, economic or seasonal factors of the place of shipment or destination change, the system can promptly adjust the relevant parameters in the forecast model to maintain the accuracy and timeliness of the forecast results;

[0059] The use of quantitative parameters and characteristic analysis methods avoids the subjectivity and uncertainty of traditional empirical forecasting methods; through scientific models and algorithms, it can objectively and accurately evaluate the impact of various factors on logistics demand, and improve the scientificity and reliability of forecasting;

[0060] By comparing the cargo demand characteristic index of the destination and the cargo supply characteristic index of the place of shipment, and selecting the characteristic index with the greatest impact on logistics demand as the logistics demand characteristic index, this method can more accurately locate the key factors affecting logistics demand; at the same time, combined with real-time cross-border transportation costs and historical data for the same period, the forecast results are corrected, further improving the precision and accuracy of the forecast;

[0061] This method provides cross-border logistics companies with more accurate and timely logistics demand forecast results, which helps them to better formulate logistics plans, optimize resource allocation, reduce operating costs and improve customer satisfaction. At the same time, it also provides strong decision-making support for companies in the face of emergencies, helping them to respond to market changes and challenges in a timely manner.

[0062] The framework and ideas of this method are sustainable and scalable; as the cross-border logistics market continues to develop and change, the prediction model can be further optimized and improved as needed, and more new influencing factors and parameters can be introduced to adapt to new market demands and challenges;

[0063] To sum up, the above-mentioned cross-border logistics demand forecasting method has obvious advantages in solving the problem of cross-border logistics demand forecasting, and can provide cross-border logistics companies with more accurate and reliable forecasting results and decision-making support. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is a flow chart of the present invention;

[0065] Figure 2 It is a flowchart of the construction method of the cross-border logistics demand impact analysis model;

[0066] Figure 3 It is the structural diagram of the cross-border logistics demand forecasting system. DETAILED DESCRIPTION

[0067] In the description of the present application, those skilled in the art should know that the present application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, the present application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, the present application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.

[0068] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0069] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws.

[0070] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.

[0071] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.

[0072] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.

[0073] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0074] The present application is described below in conjunction with the drawings in the present application.

[0075] Embodiment 1: Figure 1 to Figure 2 As shown, the cross-border logistics demand forecasting method of the present invention specifically includes the following steps:

[0076] S1. Collecting a set of parameters affecting cargo demand at the destination and a set of parameters affecting cargo supply at the place of shipment; the set of parameters affecting cargo demand at the destination includes quantitative parameters affecting policy at the destination, quantitative parameters affecting seasonality at the destination, and quantitative parameters affecting the economy at the destination;

[0077] Quantitative parameters of the impact of destination policies: by regularly reviewing official announcements, trade policy documents and information released by international economic and trade organizations of the destination country, we can identify and record policy changes related to imported goods; we can quantify policy changes for subsequent analysis;

[0078] Quantification parameters of seasonal impact of destinations: Based on historical data, analyze the changing patterns of demand for specific commodities in the destination market in different seasons; use time series analysis and seasonal decomposition methods to quantify seasonal demand changes into specific parameters;

[0079] Quantitative parameters of the destination economic impact: collect macroeconomic data of the destination country, which reflects the overall situation of the local economy and consumption capacity; use statistical analysis methods to evaluate the impact of these data on the demand for imported goods and convert them into quantifiable economic impact parameters;

[0080] The set of parameters affecting the supply of goods at the place of shipment includes quantitative parameters of policy impact at the place of shipment, quantitative parameters of seasonal impact at the place of shipment, and quantitative parameters of economic impact at the place of shipment;

[0081] Quantitative parameters of the impact of policies at the place of shipment, closely monitor the export policy dynamics of the country of shipment, including adjustments to export tax rebates, export restrictions, and export subsidy policies; similar to the quantification of the impact of policies at the destination, convert policy changes at the place of shipment into specific quantitative parameters;

[0082] Quantitative parameters of seasonal impacts at the shipping destination, analyzing changes in the shipping destination market's supply capacity for specific commodities in different seasons, taking into account the impact of natural and human factors; using methods similar to seasonal demand analysis at the destination to establish a quantitative model for seasonal supply changes;

[0083] Quantitative parameters of economic impact of the place of shipment: collect macroeconomic data of the place of shipment to assess the impact of its economic conditions on the supply capacity of export commodities; use economic analysis tools to assess how local economic conditions affect the production capacity and capital liquidity of exporters, thereby converting them into quantifiable economic impact parameters;

[0084] Data collection should cover official information, data from industry associations, market research reports, and third-party data providers;

[0085] To improve efficiency, use web crawlers and API interfaces to crawl data regularly;

[0086] Clean and pre-process the collected data to remove erroneous or irrelevant information and ensure data quality;

[0087] For different data sources, establish unified data formats and standards to facilitate subsequent data fusion and analysis.

[0088] In this step, by comprehensively considering the influencing factors of multiple dimensions such as policy, season and economy of the destination and the place of shipment, this step ensures the comprehensiveness of data collection; this multi-dimensional analysis method can more accurately reflect the complexity and dynamics of cross-border logistics demand, and provide a solid foundation for subsequent forecasts; converting influencing factors such as policy, season and economy into quantifiable parameters not only improves the accuracy of analysis, but also makes the forecast results more intuitive and easy to understand; quantitative processing allows comparison and trade-off between different influencing factors, providing a more scientific basis for decision makers; regularly checking official announcements, trade policy documents, and using web crawlers and API interfaces to capture data ensures the timeliness and freshness of data; it helps to capture market trends in a timely manner. The impact of market changes and policy adjustments on logistics demand can be improved to improve the accuracy of forecasts; through data cleaning and preprocessing steps, erroneous or irrelevant information is removed to ensure the quality and reliability of data and avoid forecast deviations caused by data quality issues; establishing a unified data format and standard will help the integration and analysis of different data sources; this standardized processing makes data easier to manage and use, and improves the efficiency and accuracy of data processing; the data and analysis results collected in this step can provide cross-border logistics companies with more comprehensive, accurate and timely logistics demand forecasts; it helps companies to formulate reasonable logistics plans, optimize resource allocation, improve operational efficiency, reduce risks, and thus maintain competitiveness in a highly competitive market environment.

[0089] S2. Perform characteristic analysis on the parameter set influencing the cargo demand at the destination and the parameter set influencing the cargo supply at the place of shipment, respectively, to obtain a characteristic index of the cargo demand at the destination and a characteristic index of the cargo supply at the place of shipment;

[0090] The method for obtaining the destination cargo demand characteristic index and the place of shipment cargo supply characteristic index comprises:

[0091] Check the parameter sets affecting the demand for goods at the destination and the supply of goods at the place of shipment and deal with errors, duplicate records and inconsistencies in the data; use interpolation methods for missing data; identify and deal with outliers through statistical methods;

[0092] Each influencing parameter is analyzed separately, using multivariate statistical analysis methods to identify the commonalities among multiple factors and extract key features;

[0093] Convert qualitative descriptions of policy impacts and seasonal changes into specific numerical indicators; standardize data from different sources so that they can be compared on the same scale;

[0094] Weights are assigned to each factor according to its importance and influence, and then the weighted average is calculated as the characteristic index.

[0095] In this step, the quality and integrity of the data are significantly improved by checking and processing errors, duplicate records and inconsistencies in the data, and using interpolation methods to supplement missing data; this provides a more reliable basis for subsequent analysis and reduces deviations caused by data problems; outliers are identified and processed by statistical methods, effectively eliminating the impact of extreme values ​​on the analysis results, making the analysis results more stable and accurate; it helps to avoid the distortion of the overall forecast results by a single extreme event; each influencing parameter is analyzed separately, and multivariate statistical analysis methods are used to identify the commonalities between multiple factors, which can fully consider the impact of various factors on logistics demand; this multi-dimensional analysis method is more comprehensive and in-depth than a single-dimensional analysis, and can more accurately reflect the actual situation; qualitative descriptions such as policy impacts and seasonal changes are converted into specific numerical indicators, so that these factors can be quantified and participate in subsequent calculations and analysis ; This transformation enhances the objectivity and operability of the analysis, making the prediction results more convincing; standardizes data from different sources so that they can be compared on the same scale, ensuring the comparability of different factors; helps to eliminate errors caused by different data dimensions and improves the accuracy of the analysis results; assigns weights to each factor according to its importance and influence, and calculates the weighted average as the characteristic index, which can more reasonably reflect the influence of each factor on logistics demand; this scientific weight allocation method avoids the interference of subjective judgment and makes the prediction results more objective and accurate; the characteristic analysis method in step S2 not only improves the quality and reliability of the data, but also provides strong support for obtaining accurate destination cargo demand characteristic index and place of shipment cargo supply characteristic index through multi-dimensional comprehensive consideration, qualitative to quantitative transformation, data standardization and scientific weight allocation.

[0096] S3. Compare the destination cargo demand characteristic index and the origin cargo supply characteristic index, extract the smallest characteristic index, and mark it as the logistics demand characteristic index;

[0097] The destination cargo demand characteristics index comprehensively reflects the impact of destination market factors on logistics demand, including quantitative assessments of policies, seasons, and economy. It represents the level of logistics demand that may be achieved from the demand side.

[0098] The cargo supply characteristics index of the place of shipment comprehensively considers the constraints of the market factors of the place of shipment on the logistics supply capacity, and also covers the quantitative evaluation of factors such as policies, seasons and economy. It represents the level that logistics supply may reach from the supply side.

[0099] In the context of cross-border logistics, logistics demand is driven not only by the demand side but also by the supply side; therefore, it is necessary to consider the characteristic indexes of these two aspects at the same time; the logic of comparison is to identify the main limiting factors of current logistics demand, that is, whether the restrictions on the demand side are more serious or the restrictions on the supply side are more prominent;

[0100] By comparing the sizes of the two characteristic indexes, the smaller one is selected as the characteristic index of logistics demand; the actual level of logistics demand is often constrained by the weakest link. If the characteristic index on the demand side is high, but the characteristic index on the supply side is low, then the actual logistics demand may be restricted due to insufficient supply;

[0101] The extracted logistics demand characteristic index will serve as the benchmark for subsequent forecasting steps.

[0102] In this step, by considering both the destination cargo demand characteristic index and the place of shipment cargo supply characteristic index, this step comprehensively evaluates the dual impacts faced by cross-border logistics demand; it helps to accurately locate the main limiting factors of current logistics demand, whether from the demand side or the supply side; by comparing the two characteristic indexes and selecting the smaller one as the logistics demand characteristic index, the "bottleneck" link in logistics demand is effectively identified; it provides an important decision-making basis for enterprises, enabling them to optimize the weakest link and improve overall logistics efficiency; the extracted logistics demand characteristic index, as a benchmark for subsequent forecasting steps, can more accurately reflect the actual level that current logistics demand may reach; it helps to reduce the errors caused by single-dimensional analysis or empirical forecasting methods, and improve the accuracy and reliability of forecasts; in the face of emergencies or market changes, it can help enterprises identify potential risks and challenges more quickly and formulate corresponding countermeasures in advance; step S3 emphasizes the requirements for data accuracy and completeness, which helps to promote the transformation of enterprises to data-driven decision-making in the field of logistics management; by continuously collecting and analyzing relevant data, enterprises can continuously optimize forecasting models and decision-making processes to improve overall operational efficiency and competitiveness.

[0103] S4. Obtain the historical freight transportation characteristic quantity corresponding to the real-time cross-border transportation cost and the logistics demand characteristic index for the same period; wherein the historical freight transportation characteristic quantity for the same period is the historical freight demand characteristic quantity of the destination for the same period or the historical freight supply characteristic quantity of the place of shipment for the same period;

[0104] The method for obtaining the real-time cross-border transportation cost includes:

[0105] Identify logistics providers as data sources; these organizations can provide real-time cross-border transportation cost information;

[0106] Use logistics information platform services, which integrate the quotation systems of multiple logistics service providers to facilitate the acquisition of comprehensive cost information;

[0107] Establish data interface with logistics information platform services to ensure real-time access to cross-border transportation cost data;

[0108] Obtain logistics parameters through logistics information platform services, including transportation routes, cargo types, weight and volume, and calculate real-time cross-border transportation costs based on logistics parameters;

[0109] Verify the real-time cross-border transportation cost data obtained and clean the noise and outliers in the data;

[0110] Store the verified real-time cross-border transportation cost data in the database for subsequent analysis and use;

[0111] The method for obtaining the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index in the same period includes:

[0112] Extracting shipping data from the company’s historical transaction records, including shipping volume, shipping frequency, shipping time, and shipping method;

[0113] Categorize data by date to facilitate comparisons over the same period;

[0114] Apply machine learning algorithms to process and analyze historical data to obtain patterns and trends in historical data, i.e., freight transportation characteristics;

[0115] By comparing the logistics demand characteristic index, select the matching historical freight transportation characteristic quantities for the same period.

[0116] In this step, by directly establishing a data interface with the logistics information platform service, cross-border transportation costs can be obtained in real time, ensuring the timeliness and accuracy of the data; at the same time, the real-time data is verified and cleaned to further improve the reliability of the data; the logistics information platform service integrates the quotation systems of multiple logistics service providers, allowing enterprises to obtain more comprehensive cross-border transportation cost information and avoid information bias that may be caused by a single data source; in addition, by considering multiple logistics parameters such as transportation routes, cargo types, weight and volume, the calculated real-time cross-border transportation costs are more accurate and diversified, meeting the needs of different scenarios; by extracting the company's historical transaction records and applying machine learning algorithms to process and analyze these data, enterprises can deeply explore patterns and trends in historical data, that is, cargo transportation characteristics; these characteristics not only reflect past market conditions, but also provide information for future logistics planning. and forecasting; at the same time, comparing the logistics demand characteristic index with the historical freight transportation characteristic quantity during the same period will help enterprises to grasp the current market demand and logistics status more accurately; the real-time cross-border transportation costs obtained in this step and the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index provide enterprises with rich data support; these data can not only help enterprises formulate more reasonable logistics budgets and plans, but also provide a strong basis for logistics strategy adjustment, supplier selection, inventory management, etc., so as to optimize the logistics operations of enterprises, reduce costs and improve competitiveness; this step realizes the intelligent processing and analysis of historical data, reduces manual intervention and errors, and improves work efficiency; at the same time, the establishment of a real-time data interface also realizes the automation of data acquisition, allowing enterprises to obtain the required information more conveniently, laying the foundation for the digital transformation and intelligent upgrading of enterprises.

[0117] S5. Taking the logistics demand characteristic index and the real-time cross-border transportation cost as inputs, inputting them into the pre-built cross-border logistics demand impact analysis model to obtain the cross-border logistics demand factor;

[0118] The method for constructing the cross-border logistics demand impact analysis model includes:

[0119] Collect historical data, including logistics demand characteristic index, real-time cross-border transportation costs and historical logistics demand in different time periods;

[0120] Clean the collected historical data, remove duplicate data, process missing values, identify and process outliers, and ensure the integrity and accuracy of the data;

[0121] Selecting machine learning models as the infrastructure of the cross-border logistics demand impact analysis model, the machine learning models include random forest, gradient boosting tree, support vector machine, neural network and recurrent neural network;

[0122] Divide the processed data into training and validation sets to evaluate the performance of the model and prevent overfitting; optimize the performance of the model on the validation set by adjusting the model parameters; during the model training process, understand which factors have the most significant impact on logistics demand through feature importance evaluation; help understand the decision logic of the model and provide guidance for subsequent data collection and feature engineering;

[0123] The prediction performance of the model was evaluated using the validation set; the evaluation indicators included mean square error, mean absolute error, and root mean square error;

[0124] Adjust the model structure, parameters, and feature engineering scheme based on the evaluation results to improve the model’s predictive performance;

[0125] Deploy the trained model to the production environment to achieve real-time or periodic logistics demand forecasting;

[0126] The mathematical formula for calculating the cross-border logistics demand factor in the cross-border logistics demand impact analysis model is:

[0127] F=w 1 I+w 2 ·C+b

[0128] Among them, F represents the cross-border logistics demand factor, I represents the logistics demand characteristic index, C represents the real-time cross-border transportation cost, and w 1 represents the impact of the logistics demand characteristic index on the cross-border logistics demand factor, w 2 It represents the influence of real-time cross-border transportation cost on cross-border logistics demand factor, and b represents the bias term, which is used to adjust the baseline prediction value of the model.

[0129] In this step, by collecting and processing historical data and combining it with advanced machine learning models, the model can capture complex nonlinear relationships and thus more accurately predict cross-border logistics demand. This helps companies make more accurate business decisions and optimize resource allocation. Feature importance evaluation during model training not only helps understand which factors have the most significant impact on logistics demand, but also enhances the interpretability of the model. This enables companies to deeply understand the decision logic of the model and provide strong support for subsequent business analysis and strategy adjustments. By dividing the training set and validation set, and using the validation set to evaluate and adjust model parameters, overfitting can be effectively prevented and the model's generalization ability on unknown data can be improved. In addition, the continuous model evaluation and optimization process ensures that the model can be used as needed. It continuously improves with the changes in time and data, maintaining the stability and superiority of its forecasting performance; accurate logistics demand forecasting enables enterprises to plan logistics resources and transportation plans in advance, thereby improving the flexibility and responsiveness of the supply chain; in the face of market fluctuations and emergencies, enterprises can adjust strategies more quickly, reduce risks and seize opportunities; by optimizing logistics demand and resource allocation, enterprises can reduce unnecessary inventory backlogs and transportation costs, improve operational efficiency and reduce overall operating costs; the deployment of cross-border logistics demand impact analysis models realizes real-time or regular forecasting of logistics demand, providing enterprises with strong data support; this enables enterprises to guide business operations and development directions based on data-driven intelligent decision-making, and promote the digital transformation and intelligent upgrading of enterprises.

[0130] S6. Based on the cross-border logistics demand factor, the historical freight transportation characteristic quantity is corrected to obtain the cross-border logistics demand forecast;

[0131] The method for obtaining the cross-border logistics demand forecast includes:

[0132] According to the specific value of the cross-border logistics demand factor, combined with historical data and business experience, the correction coefficient is set; the correction coefficient is intended to adjust the characteristic quantity of freight transportation in the same period of history to a level that is more in line with the current market environment and cost conditions;

[0133] Select the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index as the benchmark; this quantity can be the historical freight demand characteristic quantity of the destination during the same period, or the historical freight supply characteristic quantity of the place of shipment during the same period, depending on the source and nature of the logistics demand characteristic index;

[0134] Apply the correction coefficient to adjust the selected historical freight transportation characteristics in the same period to obtain the corrected logistics demand forecast; the corrected data can reflect the reasonable forecast value of cross-border logistics demand under the current policy, season, economy and transportation cost conditions;

[0135] Use the latest actual logistics data to verify the corrected logistics demand forecast and evaluate the accuracy of the forecast;

[0136] The verified and adjusted cross-border logistics demand forecast will be used as the final output and as an important basis for enterprises to formulate logistics plans, allocate resources and optimize costs.

[0137] In this step, by introducing the cross-border logistics demand factor, the impact of various market factors and real-time cross-border transportation costs on logistics demand can be comprehensively considered, thereby improving the accuracy of the forecast; this comprehensive consideration method can better reflect the complexity of the current market environment than single-dimensional historical data analysis; setting correction coefficients and adjusting them based on real-time data allows the forecast method to flexibly respond to sudden changes in the market environment. This flexibility ensures that the forecast results can quickly reflect the actual market situation and provide strong support for the company's decision-making; accurate logistics demand forecasts can help companies better formulate logistics plans and allocate resources reasonably; companies can arrange transportation tools, storage facilities and human resources in advance based on forecast results, thereby avoiding resource waste and shortages and improving operational efficiency; by optimizing resource allocation and reducing unnecessary logistics activities By doing so, enterprises can reduce operating costs; accurate logistics demand forecasting can help enterprises reduce inventory backlogs, shorten transportation cycles, reduce transportation costs, etc., thereby improving overall profitability; accurate logistics demand forecasting can ensure that enterprises meet customer needs on time and in quantity, thereby improving customer satisfaction; good customer service experience helps to enhance customer loyalty and win more market share for enterprises; enterprises with efficient logistics demand forecasting capabilities have more advantages in market competition; they can respond to market changes more quickly, seize business opportunities, and reduce operational risks, thereby standing out in the industry; the cross-border logistics demand forecasting method in step S6 not only improves the accuracy of the forecast, but also enhances the enterprise's market adaptability, resource allocation efficiency, cost control capabilities and customer satisfaction, which ultimately helps to improve the overall competitiveness and market position of the enterprise.

[0138] Embodiment 2: Figure 3 As shown, the cross-border logistics demand forecasting system of the present invention specifically includes the following modules:

[0139] Data collection module, collecting the parameter set affecting the cargo demand at the destination and the parameter set affecting the cargo supply at the place of shipment;

[0140] The characteristic analysis module performs characteristic analysis on the parameter set affecting the demand of the destination goods and the parameter set affecting the supply of the goods at the place of shipment, respectively, to obtain the characteristic index of the demand of the destination goods and the characteristic index of the supply of the goods at the place of shipment;

[0141] The index comparison module compares the destination cargo demand characteristic index and the origin cargo supply characteristic index, extracts the smallest characteristic index, and marks it as the logistics demand characteristic index;

[0142] The historical data association module obtains the real-time cross-border transportation cost and the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index;

[0143] The demand impact analysis module takes the logistics demand characteristic index and the real-time cross-border transportation cost as inputs into the pre-built cross-border logistics demand impact analysis model to obtain the cross-border logistics demand factor;

[0144] The demand forecast correction module corrects the historical freight transportation characteristic quantities in the same period based on the cross-border logistics demand factors to obtain the cross-border logistics demand forecast.

[0145] The system collects quantitative parameters including destination and shipping place policies, seasons, economy, etc. through the data collection module, achieving comprehensive coverage of factors affecting logistics demand. The integration of this multi-dimensional data enables the forecasting model to consider various complex factors more comprehensively, thereby improving the accuracy and reliability of the forecast.

[0146] The system uses the feature analysis module and index comparison module to analyze and compare the feature indexes of the destination and the place of shipment in real time, extracting the factors that have the greatest impact on logistics demand; enabling the system to respond quickly to market changes and policy adjustments, and improving the timeliness and flexibility of forecasting;

[0147] The historical data association module combines the real-time cross-border transportation cost with the logistics demand characteristic index and searches for relevant historical data for the same period as a reference. It not only utilizes the empirical value of historical data, but also corrects it with real-time data, making the prediction results closer to the current actual situation.

[0148] The demand impact analysis module uses a pre-built cross-border logistics demand impact analysis model, takes the logistics demand characteristic index and real-time cross-border transportation cost as input, and calculates the cross-border logistics demand factor through the model. Compared with the traditional empirical method, it can more accurately reflect the impact of various factors on logistics demand;

[0149] The demand forecast correction module corrects the historical freight transportation characteristics based on the cross-border logistics demand factor, and finally obtains the cross-border logistics demand forecast; ensuring that the forecast results take into account both historical trends and current actual conditions, thereby improving the accuracy and practicality of the forecast results;

[0150] The entire system achieves rapid and accurate prediction of cross-border logistics demand through automation and intelligence; this helps enterprises plan logistics resources in advance, optimize logistics layout, improve logistics efficiency, and reduce operating costs. It also helps enterprises better respond to market changes and risk challenges.

[0151] In summary, by integrating various data sources and applying advanced analytical techniques, the system can significantly improve the accuracy and reliability of cross-border logistics demand forecasts, helping companies maintain their competitiveness in the complex and changing international trade environment.

[0152] The various variations and specific embodiments of the cross-border logistics demand forecasting method in the aforementioned embodiment 1 are also applicable to the cross-border logistics demand forecasting system of this embodiment. Through the aforementioned detailed description of the cross-border logistics demand forecasting method, those skilled in the art can clearly understand the implementation method of the cross-border logistics demand forecasting system in this embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here.

[0153] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0154] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cross-border logistics demand forecasting method, characterized in that: The method comprises: Collect the parameter set affecting the cargo demand at the destination and the parameter set affecting the cargo supply at the place of shipment; Perform characteristic analysis on the parameter set influencing the cargo demand at the destination and the parameter set influencing the cargo supply at the place of shipment respectively, and obtain the characteristic index of the cargo demand at the destination and the characteristic index of the cargo supply at the place of shipment; Compare the destination cargo demand characteristic index and the origin cargo supply characteristic index, extract the smallest characteristic index, and mark it as the logistics demand characteristic index; Obtain the real-time cross-border transportation cost and the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index; The logistics demand characteristic index and real-time cross-border transportation cost are used as inputs into the pre-built cross-border logistics demand impact analysis model to obtain the cross-border logistics demand factor; Based on the cross-border logistics demand factor, the characteristic quantities of freight transportation in the same period in history are corrected to obtain the forecast quantity of cross-border logistics demand.

2. The cross-border logistics demand forecasting method according to claim 1, characterized in that: The mathematical formula for calculating the cross-border logistics demand factor in the cross-border logistics demand impact analysis model is: F=w1·I+w2·C+b Among them, F represents the cross-border logistics demand factor, I represents the logistics demand characteristic index, C represents the real-time cross-border transportation cost, w1 represents the influence of the logistics demand characteristic index on the cross-border logistics demand factor, w2 represents the influence of the real-time cross-border transportation cost on the cross-border logistics demand factor, and b represents the bias term, which is used to adjust the baseline prediction value of the model.

3. The cross-border logistics demand forecasting method according to claim 2, characterized in that: The method for constructing the cross-border logistics demand impact analysis model includes: Collect historical data, including logistics demand characteristic index, real-time cross-border transportation costs and historical logistics demand in different time periods; Clean the collected historical data, remove duplicate data, process missing values, and identify and process outliers; Selecting machine learning models as the infrastructure of the cross-border logistics demand impact analysis model, the machine learning models include random forest, gradient boosting tree, support vector machine, neural network and recurrent neural network; Divide the processed data into a training set and a validation set, and train the selected model using the training set; Optimize the performance of the model on the validation set by adjusting the model parameters; Use an independent test set to evaluate the model's predictive performance and adjust the model structure, parameters, and feature engineering scheme based on the evaluation results; Deploy the trained model to production environment.

4. The cross-border logistics demand forecasting method according to claim 1, characterized in that: The method for obtaining the destination cargo demand characteristic index and the place of shipment cargo supply characteristic index comprises: Check the parameter sets affecting the demand for goods at the destination and the parameter sets affecting the supply of goods at the origin and deal with errors, duplicate records and inconsistencies in the data; Analyze each influencing parameter separately, identify the commonalities among multiple factors, and extract key features; Convert qualitative descriptions of policy impacts and seasonal changes into concrete numerical indicators; Weights are assigned to each factor according to its importance and influence, and then the weighted average is calculated as the characteristic index.

5. The cross-border logistics demand forecasting method according to claim 1, characterized in that: The method for obtaining the real-time cross-border transportation cost includes: Identify logistics providers as data sources; Use logistics information platform services to obtain cost information; Establish data interface with logistics information platform service; Obtain logistics parameters through logistics information platform services and calculate real-time cross-border transportation costs based on logistics parameters; Verify the real-time cross-border transportation cost data obtained and clean the noise and outliers in the data; The verified real-time cross-border transportation cost data is stored in the database.

6. The cross-border logistics demand forecasting method according to claim 1, characterized in that: The method for obtaining the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index in the same period includes: Extracting shipping data from the company’s historical transaction records; Categorize data by date to allow for year-over-year comparisons; Process and analyze historical data to obtain patterns and trends in historical data, i.e., freight transportation characteristics; By comparing the logistics demand characteristic index, select the matching historical freight transportation characteristic quantities for the same period.

7. The cross-border logistics demand forecasting method according to claim 1, characterized in that: The method for obtaining the cross-border logistics demand forecast includes: Set the correction coefficient according to the specific value of the cross-border logistics demand factor; Select the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index as the benchmark; Apply the correction coefficient to adjust the selected historical freight transportation characteristic quantity in the same period to obtain the corrected logistics demand forecast quantity; Use the latest actual logistics data to verify the corrected logistics demand forecast and evaluate the accuracy of the forecast; The verified and adjusted cross-border logistics demand forecast is used as the final output.

8. A cross-border logistics demand forecasting system, characterized in that: The system comprises: Data collection module, collecting the parameter set affecting the cargo demand at the destination and the parameter set affecting the cargo supply at the place of shipment; The characteristic analysis module performs characteristic analysis on the parameter set affecting the demand of the destination goods and the parameter set affecting the supply of the goods at the place of shipment, respectively, to obtain the characteristic index of the demand of the destination goods and the characteristic index of the supply of the goods at the place of shipment; The index comparison module compares the destination cargo demand characteristic index and the origin cargo supply characteristic index, extracts the smallest characteristic index, and marks it as the logistics demand characteristic index; The historical data association module obtains the real-time cross-border transportation cost and the historical freight transportation characteristic quantity corresponding to the logistics demand characteristic index; The demand impact analysis module takes the logistics demand characteristic index and the real-time cross-border transportation cost as inputs into the pre-built cross-border logistics demand impact analysis model to obtain the cross-border logistics demand factor; The demand forecast correction module corrects the historical freight transportation characteristic quantities in the same period based on the cross-border logistics demand factors to obtain the cross-border logistics demand forecast.

9. An electronic device for cross-border logistics demand forecasting, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

Images