A method for constructing a transaction decision model for logistics data analysis

Abstract

The present invention relates to the field of model construction technology, and in particular to a method for constructing a transaction decision model for logistics data analysis. The method comprises the following steps: obtaining comprehensive logistics data; performing data preprocessing on the comprehensive logistics data to generate standard comprehensive logistics data; performing cargo value assessment on the standard comprehensive logistics data to generate cargo value assessment data; performing initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate initial logistics transportation mode data; performing first logistics transaction fee confirmation on the initial logistics transportation mode data based on the cargo value assessment data to obtain first logistics transportation transaction data; performing intelligent perception of logistics transportation cargo using sensors to obtain logistics cargo intelligent perception data. The present invention improves the accuracy and comprehensiveness of transaction decision model construction through standardized data processing, dynamic transportation mode selection, reverse logistics cost analysis and machine learning decision making.

Description

Technical Field

[0001] The present invention relates to the technical field of model construction, and in particular to a method for constructing a transaction decision model for logistics data analysis. Background Art

[0002] In the early days, logistics management mainly relied on traditional linear programming and operations research methods. Although these methods can effectively solve basic transportation and inventory problems, they are unable to cope with large-scale and complex supply chain data. With the advancement of information technology, especially the rise of data mining and machine learning, logistics data analysis has gradually shifted to data-driven methods. In the 1990s, enterprises introduced decision support systems (DSS) based on data warehouses, which optimized transportation routes, inventory management and other links by storing and analyzing historical data. Entering the 21st century, the rapid development of cloud computing and big data technologies has enabled enterprises to process massive amounts of data and optimize decisions using real-time data analysis. In recent years, the application of deep learning and artificial intelligence has further promoted the progress of logistics data analysis. Advanced algorithms, such as neural networks and reinforcement learning, are used to predict demand, optimize inventory and improve logistics efficiency. These technologies can provide more accurate decision support by identifying and predicting complex data patterns. However, traditional logistics transactions currently do not fully consider the lack of cargo value on transportation mode selection and cost calculation, and the problem of insufficient estimation of return and reverse logistics costs, which leads to low accuracy and comprehensiveness in the construction of transaction decision models. Summary of the invention

[0003] Based on this, it is necessary to provide a method for constructing a transaction decision model for logistics data analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for constructing a transaction decision model for logistics data analysis is provided, the method comprising the following steps:

[0005] Step S1: Acquire comprehensive logistics data; perform data preprocessing on the comprehensive logistics data to generate standard comprehensive logistics data; perform cargo value assessment on the standard comprehensive logistics data to generate cargo value assessment data; perform initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate initial logistics transportation mode data;

[0006] Step S2: based on the cargo value assessment data, the first logistics transaction fee of the initial logistics transportation mode data is confirmed to obtain the first logistics transportation transaction data; the logistics transport cargo is intelligently sensed by the sensor to obtain the logistics cargo intelligent sensing data; the logistics cargo intelligent sensing data is dynamically selected to select the transportation mode to generate the dynamic transportation mode data; the transaction fee of the first logistics transportation transaction data is adjusted by the dynamic transportation mode data to generate the second logistics transportation transaction data;

[0007] Step S3: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data; reverse logistics cost analysis is performed on the cargo logistics delivery feedback data to generate reverse logistics cost data; transaction cost adjustment is performed on the first logistics transportation transaction data based on the reverse logistics cost data to generate third logistics transportation transaction data;

[0008] Step S4: construct transaction adjustment rules based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data; use machine learning methods to make transaction cost decisions on the first logistics and transportation transaction data based on the transaction adjustment rule data, thereby generating a transaction cost decision report.

[0009] The present invention eliminates noise and inconsistency in the data by preprocessing and standardizing the comprehensive logistics data, thereby ensuring the quality and consistency of the data. This makes subsequent analysis more accurate, reduces errors and deviations caused by data inconsistency, and thus improves the reliability of the overall decision. The value assessment of the goods ensures that the transportation decision can fully consider the economic importance of the goods. This enables high-value goods to be given priority when selecting transportation modes and allocating resources, optimizes resource allocation, and improves transportation efficiency and cost-effectiveness. Based on the initial transportation mode analysis of the goods value assessment data, a transportation plan that better meets the needs of the goods can be formulated. This provides a scientific basis for subsequent cost confirmation, helps to formulate a reasonable transportation plan and budget, and thus improves the accuracy of transportation decisions. By preliminarily confirming the logistics transaction costs, a preliminary basis is provided for budget formulation and cost control during the transportation process. This helps to avoid budget overruns, ensure that costs are controlled within an acceptable range, and optimize financial planning. By using sensor technology for intelligent perception, the status and environmental information of logistics goods can be obtained in real time, and the transportation mode can be dynamically adjusted. This flexibility improves transportation efficiency, can respond to changes in actual conditions in a timely manner, and reduce delays and losses. Adjusting initial shipping costs based on dynamic shipping mode data makes cost calculations more accurate and realistic. Adjusting costs based on real-time data can better reflect the actual costs during transportation and reduce problems caused by inaccurate cost estimates in static modes. Providing real-time feedback during transportation can promptly identify and resolve problems such as shipping delays and damage. This helps improve the quality of transportation services, increase customer satisfaction, and reduce additional costs caused by logistics problems. Reverse logistics cost analysis can comprehensively evaluate the costs of returns and reverse logistics, ensuring that all relevant costs are taken into account. This makes cost estimates more comprehensive and avoids budget shortfalls caused by ignoring reverse logistics costs. Adjusting costs based on reverse logistics cost data ensures that the final cost calculation is more accurate. By comprehensively considering all relevant costs, budget management is optimized and financial losses caused by inadequate consideration of reverse logistics costs are reduced. Building transaction adjustment rules can standardize the cost adjustment process and improve the consistency and repeatability of decisions. This reduces decision inconsistency caused by the lack of clear rules and ensures the transparency and fairness of the transaction cost adjustment process. Using machine learning methods to learn from transaction adjustment rule data, the decision process is optimized and the level of intelligence in decision making is improved. By analyzing historical data, an accurate cost decision report is generated, thereby enhancing the scientificity and accuracy of decision-making and further optimizing logistics cost management. Therefore, the present invention improves the accuracy and comprehensiveness of transaction decision model construction through standardized data processing, dynamic transportation mode selection, reverse logistics cost analysis and machine learning decision-making.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Obtaining comprehensive logistics data;

[0012] Step S12: preprocessing the logistics comprehensive data to generate standard logistics comprehensive data, wherein the data preprocessing includes data cleaning, data missing value filling and data standardization;

[0013] Step S13: screening the transported goods data from the standard logistics comprehensive data to obtain logistics transported goods information data;

[0014] Step S14: perform cargo value assessment on the logistics transportation cargo information data to generate cargo value assessment data; perform initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate logistics initial transportation mode data, wherein the logistics initial transportation mode data includes air transportation mode data, sea transportation mode data and land transportation mode data.

[0015] The present invention not only improves the consistency and integrity of the data, but also provides a reliable basis for subsequent analysis by preprocessing the acquired comprehensive logistics data to ensure data cleaning, missing value filling and standardization. By screening the standard comprehensive logistics data, the logistics transportation cargo information data is accurately obtained. This helps to accurately locate the goods involved in the logistics process and improve the efficiency of logistics management. Cargo value assessment is performed on the logistics transportation cargo information data to generate cargo value assessment data, which can provide important economic references for logistics management and make resource allocation more reasonable. Based on the cargo value assessment data, preliminary analysis and generation of logistics initial transportation mode data including air transportation, sea transportation and land transportation are performed. Through these data, the optimal mode of transportation can be selected to improve the efficiency of logistics transportation and reduce transportation costs. The generation of logistics initial transportation mode data covers multiple modes of air transportation, sea transportation and land transportation, and can select the most suitable mode of transportation according to actual needs, thereby improving the flexibility and adaptability of logistics transportation.

[0016] Preferably, step S14 comprises the following steps:

[0017] Step S141: Analyze the characteristics of the goods transported by the logistics transport information data to generate goods characteristic data;

[0018] Step S142: performing market price trend analysis of similar items on the logistics transport cargo information data based on the cargo characteristic data to generate cargo market trend data;

[0019] Step S143: Performing a comprehensive value assessment of the goods through the goods market trend data and the goods characteristic data to generate goods value assessment data, wherein the comprehensive value assessment of the goods includes cost estimation and market value analysis;

[0020] Step S144: Performing a transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate logistics transportation mode data.

[0021] The present invention generates goods characteristic data by analyzing the characteristics of the information data of the goods transported by the logistics. This step ensures a clear understanding of the characteristics of the goods, which is helpful for the subsequent market analysis and value assessment. The market price trends of similar items are analyzed to generate goods market trend data. This analysis can provide market dynamic information, so that logistics managers can understand market demand and price fluctuations and make more forward-looking decisions. By combining the goods market trend data and the goods characteristic data, a comprehensive value assessment is performed, including cost estimation and market value analysis. Such a comprehensive assessment method can accurately reflect the true value of the goods and help logistics companies optimize resource allocation and pricing strategies. The transportation mode analysis is performed on the information data of the goods transported by the logistics to generate logistics transportation mode data. By evaluating the comprehensive value of the goods, the most suitable transportation mode can be selected, thereby improving logistics efficiency and reducing transportation costs. The entire process of step S14 provides data-based decision support for logistics management through systematic data analysis and evaluation, ensuring the scientificity and rationality of the decision. Through a comprehensive analysis of the characteristics, market trends and comprehensive value of the goods, the economic benefits of logistics management can be improved, and the economy and sustainability of logistics transportation can be ensured.

[0022] Preferably, step S144 includes the following steps:

[0023] Step S1441: Locate the starting point and the end point of the transportation of the goods in the logistics transportation, and obtain the starting point data and the end point data of the transportation of the goods;

[0024] Step S1442: performing cargo characteristic demand analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate cargo characteristic demand data;

[0025] Step S1443: Evaluate the transportation mode of the cargo transportation starting point data and the cargo transportation destination data according to the cargo characteristic demand data to generate logistics initial transportation mode data, wherein the logistics initial transportation mode data includes air transportation mode data, sea transportation mode data and land transportation mode data.

[0026] The present invention obtains the cargo transportation starting point data and cargo transportation end point data by locating the transportation starting point and end point of the logistics transportation cargo information data. This ensures the accuracy and feasibility of the transportation plan and helps to optimize the logistics path. The cargo characteristics demand analysis is performed on the logistics transportation cargo information data based on the cargo value assessment data to generate cargo characteristics demand data. This step can provide a customized transportation plan by analyzing the characteristics demand of the cargo, and improve the safety and efficiency of cargo transportation. The transportation mode is evaluated for the cargo transportation starting point data and the cargo transportation end point data according to the cargo characteristics demand data, and the logistics initial transportation mode data is generated. Through this evaluation, the most suitable transportation mode can be selected, including air transportation, sea transportation and land transportation, to ensure the efficiency and economy of the transportation process. The generation of the logistics initial transportation mode data covers a variety of transportation modes. In this way, the transportation mode can be flexibly selected according to the characteristics of the goods and the transportation demand, and the flexibility and adaptability of the transportation can be improved. Through detailed positioning and demand analysis, the transportation plan can be optimized, the transportation cost can be reduced, and the overall efficiency and benefits of logistics transportation can be improved. The transportation mode evaluation based on the cargo characteristics demand data can select the most suitable transportation mode, reduce the risk in the transportation process, and ensure that the goods arrive at the destination safely.

[0027] Preferably, step S2 comprises the following steps:

[0028] Step S21: Perform logistics transportation path planning based on the logistics initial transportation mode data to generate logistics transportation path planning data; use the goods value assessment data to confirm the first logistics transaction fee of the logistics transportation path planning data to obtain the first logistics transportation transaction data;

[0029] Step S22: Using sensors to perform intelligent perception of logistics transport goods to obtain intelligent perception data of logistics goods; building a sensor network for the intelligent perception data of logistics goods to obtain a transport goods sensor network; collecting real-time cargo transportation information on the transport goods sensor network to obtain real-time cargo transportation data;

[0030] Step S23: performing edge computing on the real-time data of cargo transportation to generate cargo transportation edge computing data; performing dynamic transportation mode selection on the cargo transportation edge computing data to generate dynamic transportation mode data, wherein the dynamic transportation mode data includes a single transportation mode and a multi-mode transportation mode;

[0031] Step S24: When it is confirmed that the dynamic transportation mode data is a multi-modal transportation mode, the transaction fee of the first logistics transportation transaction data is adjusted based on the air transportation mode data, the sea transportation mode data and the land transportation mode data to generate the second logistics transportation transaction data; when it is confirmed that the dynamic transportation mode data is a single transportation mode, the first logistics transportation transaction data is maintained unchanged.

[0032] The present invention generates logistics transportation path planning data by performing logistics transportation path planning based on the initial logistics transportation mode data. This ensures the optimization of logistics transportation paths, improves transportation efficiency, and uses the goods value assessment data to confirm the first logistics transaction fee, obtains the first logistics transportation transaction data, and ensures the rationality and transparency of the fee. The logistics transport goods are intelligently sensed by sensors to obtain logistics goods intelligent sensing data, and a transportation goods sensor network is built. Through real-time cargo transportation information collection, real-time cargo transportation data is obtained. This provides real-time monitoring and tracking during cargo transportation to ensure the safety and controllability of cargo transportation. By performing edge computing on the real-time cargo transportation data, cargo transportation edge computing data is generated, and dynamic transportation mode selection is performed to generate dynamic transportation mode data. This step improves the efficiency and real-time performance of data processing through edge computing, dynamically selects the optimal transportation mode according to the real-time situation, and improves transportation flexibility and response speed. When the dynamic transportation mode data is confirmed to be a multi-mode transportation mode, the transaction fee of the first logistics transportation transaction data is adjusted according to different transportation modes (air transportation, sea transportation, and land transportation) to generate second logistics transportation transaction data. This ensures the accuracy and rationality of cost calculation under the multi-mode transportation mode. If the dynamic transportation mode data is confirmed to be a single transportation mode, the first logistics transportation transaction data is maintained unchanged, simplifying the process and improving efficiency. Through intelligent perception, real-time data collection, edge computing and dynamic transportation mode selection, the entire transportation process becomes more efficient and safer. This not only improves the reliability of cargo transportation, but also reduces transportation time and cost. Data-driven path planning, intelligent perception and edge computing provide scientific decision-making support for logistics management, ensuring that every step in the transportation process is supported by data, improving the scientificity and effectiveness of decision-making. Dynamic transportation mode selection and cost adjustment mechanism enable logistics transportation to flexibly respond to different transportation needs and emergencies, and improve the adaptability and flexibility of the logistics system.

[0033] Preferably, performing dynamic transportation mode selection on the cargo transportation edge computing data to generate dynamic transportation mode data includes:

[0034] Perform environmental data analysis on the edge computing data of cargo transportation to generate cargo transportation environmental data, where the environmental data analysis includes temperature analysis, humidity analysis, and vibration analysis;

[0035] Cargo status evaluation is performed on cargo transportation environment data to generate cargo status evaluation data; conditional judgment is performed on cargo status evaluation data based on preset transportation conditions, and when the cargo status evaluation data does not meet the preset transportation conditions, transportation mode switching is performed based on the cargo status evaluation data to generate dynamic transportation mode data;

[0036] Among them, the transportation mode switching based on cargo status assessment data includes:

[0037] Perform environmental anomaly identification on the cargo transportation environmental data to obtain abnormal cargo transportation environmental data; when the temperature in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the sea transportation mode data; when the humidity in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the land transportation mode data; when the vibration in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the sea transportation mode data.

[0038] The present invention generates cargo transportation environment data by analyzing environmental data such as temperature, humidity and vibration of cargo transportation edge computing data, ensuring real-time monitoring of the environmental conditions of the cargo during transportation. By evaluating the cargo status of the cargo transportation environment data, the cargo status evaluation data is generated to ensure accurate understanding of the cargo status. This helps to timely discover and solve potential problems during transportation and ensure the safety of the cargo. The cargo status evaluation data is conditionally judged based on the preset transportation conditions. When the cargo status evaluation data does not meet the preset transportation conditions, the transportation mode is switched to generate dynamic transportation mode data. Such a dynamic adjustment mechanism ensures the safety and integrity of the cargo during transportation. By identifying environmental anomalies of the cargo transportation environment data, abnormal conditions such as temperature, humidity and vibration can be discovered in time, and corresponding transportation mode switching measures can be taken to ensure the safe transportation of the cargo in different environments. When the temperature, humidity and vibration are confirmed to be abnormal, the air transport, sea transport and land transport modes are switched respectively. Such a flexible response mechanism improves the flexibility and safety of the transportation process, ensuring that the cargo can arrive at the destination safely under different environmental conditions. By real-time monitoring and dynamic adjustment of transportation modes, we can effectively reduce risks in the transportation process, reduce cargo damage caused by environmental changes, and improve customer satisfaction and the credibility of logistics companies. When encountering environmental abnormalities, timely switching to the most suitable transportation mode can avoid transportation delays and losses, optimize transportation efficiency, and reduce transportation costs. The entire dynamic transportation mode selection process is based on real-time environmental data and cargo status assessment data, providing scientific data support to ensure the scientificity and accuracy of transportation decisions.

[0039] Preferably, step S3 comprises the following steps:

[0040] Step S31: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data;

[0041] Step S32: evaluating the feedback satisfaction of the cargo logistics delivery feedback data to generate cargo logistics delivery satisfaction data;

[0042] Step S33: performing reverse logistics cost analysis on the cargo logistics delivery feedback data through the cargo logistics delivery satisfaction data to generate reverse logistics cost data;

[0043] Step S34: adjusting the transaction costs of the first logistics and transportation transaction data based on the reverse logistics cost data to generate third logistics and transportation transaction data.

[0044] The present invention obtains the feedback data of cargo logistics delivery by performing cargo transportation delivery feedback according to the real-time data of cargo transportation. This step ensures the timely feedback of the cargo after delivery, which helps to quickly understand and solve potential problems. The feedback satisfaction evaluation is performed on the feedback data of cargo logistics delivery to generate the satisfaction data of cargo logistics delivery. This step helps to evaluate the customer satisfaction with the logistics service and provides an important indicator for measuring the quality of logistics services. The reverse logistics cost analysis is performed on the feedback data of cargo logistics delivery through the satisfaction data of cargo logistics delivery to generate the reverse logistics cost data. This step helps to understand the cost of reverse logistics links such as returns and exchanges, and provides data support for optimizing logistics costs. The transaction fee of the first logistics transportation transaction data is adjusted based on the reverse logistics cost data to generate the third logistics transportation transaction data. By adjusting the transaction fee, the actual logistics cost can be more accurately reflected, and the accuracy and rationality of the cost calculation can be improved. The reverse logistics cost analysis helps to identify and reduce the cost of the reverse logistics link, optimize the overall logistics cost, and improve the logistics efficiency. Through real-time data feedback and satisfaction evaluation, more accurate and timely data support can be provided to help enterprises make more scientific and precise decisions. Adjusting transaction costs based on reverse logistics cost data can flexibly respond to changes in the actual transportation process and ensure the rationality and flexibility of cost calculation. Through timely feedback and satisfaction evaluation, problems in logistics services can be discovered and improved, operational efficiency can be improved, and delays and errors in logistics links can be reduced.

[0045] Preferably, step S33 includes the following steps:

[0046] Step S331: performing feedback score matching on the cargo logistics delivery satisfaction data and the cargo logistics delivery feedback data to generate feedback score matching data; performing correlation analysis on the feedback score matching data to generate delivery satisfaction influencing factor data;

[0047] Step S332: classify the goods feedback on the goods logistics delivery feedback data to generate positive feedback goods data and negative feedback goods data; perform return frequency analysis on the negative feedback goods data to generate return frequency analysis data;

[0048] Step S333: Based on the delivery satisfaction influencing factor data, the return frequency analysis data and the positive feedback goods data are used to identify the reverse logistics cost elements, and generate reverse logistics cost element data;

[0049] Step S334: Perform reverse logistics cost analysis on the goods logistics delivery feedback data through reverse logistics cost element data to generate reverse logistics cost data, wherein the reverse logistics cost analysis includes total cost summary, cost structure analysis and trend analysis.

[0050] The present invention generates feedback score matching data by matching the satisfaction data of goods logistics delivery and the feedback data of goods logistics delivery, and performs correlation analysis to generate data of factors affecting delivery satisfaction. This step helps to accurately understand the main factors affecting customer satisfaction, thereby providing targeted guidance for improving logistics services. By classifying the feedback data of goods logistics delivery, positive feedback goods data and negative feedback goods data are generated, and return frequency analysis is performed to generate return frequency analysis data. This step helps to clarify the specific circumstances of positive and negative feedback, and provides basic data for further reverse logistics cost analysis. Based on the data of factors affecting delivery satisfaction, the return frequency analysis data and the positive feedback goods data are identified as reverse logistics cost elements, and reverse logistics cost element data are generated. This step helps to identify various cost elements in the reverse logistics process, and provides a basis for accurately calculating the reverse logistics cost. The reverse logistics cost element data is used to perform reverse logistics cost analysis on the feedback data of goods logistics delivery, and reverse logistics cost data is generated, including total cost summary, cost structure analysis and trend analysis. This step helps to fully understand the cost structure and change trend of reverse logistics, and provides decision support for logistics cost control and optimization. Through precise feedback score matching and correlation analysis, key factors affecting customer satisfaction can be identified, logistics services can be improved in a targeted manner, and customer satisfaction and loyalty can be improved. Through detailed cargo feedback classification and return frequency analysis, the reverse logistics process can be better managed, the return rate can be reduced, and logistics efficiency can be improved. Comprehensive identification of reverse logistics cost elements and detailed cost analysis can help identify and control various costs in the reverse logistics process, thereby reducing overall logistics costs and improving profitability.

[0051] Preferably, step S4 comprises the following steps:

[0052] Step S41: constructing a transaction adjustment rule based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data;

[0053] Step S42: constructing a transaction decision model through the transaction adjustment rule data to obtain a transaction decision model; using the transaction decision model to make a transaction cost decision on the first logistics transportation transaction data, thereby generating transaction cost decision data;

[0054] Step S43: Visualize the transaction fee decision data to generate a transaction fee decision report.

[0055] The present invention generates transaction adjustment rule data by constructing transaction adjustment rules based on the third logistics and transportation transaction data and the second logistics and transportation transaction data. This helps to formulate scientific and reasonable transaction adjustment rules and ensure the accuracy and fairness of transaction decisions. The transaction adjustment rule data is used to construct a transaction decision model to generate a transaction decision model. This model provides systematic and data-driven support for transaction decisions and helps to make more accurate transaction cost decisions. By using the transaction decision model to make transaction cost decisions on the first logistics and transportation transaction data, transaction cost decision data is generated. This step ensures the scientificity and rationality of transaction cost decisions and helps to improve the accuracy and transparency of costs. The establishment and application of the transaction decision model provides data-based decision support, helping enterprises to make more accurate and optimized decisions in the transaction process, and improving the scientificity and effectiveness of decisions. By visualizing the transaction cost decision data and generating a transaction cost decision report, the transparency of the transaction decision is enhanced. The visualized report makes the decision process and results more intuitive and easy to understand, and is convenient for display and communication to relevant parties. By generating a transaction cost decision report, support can be provided for the strategic decision of the enterprise, helping the management understand the basis and impact of the cost decision, so as to make more favorable strategic decisions. Through scientific cost decision-making and visual analysis, we can better manage and control logistics and transportation costs, identify and optimize cost structures, and thus improve the company's cost control capabilities and profitability.

[0056] Preferably, step S42 includes the following steps:

[0057] Step S421: dividing the transaction adjustment rule data into data sets to obtain a model training set and a model test set;

[0058] Step S422: Perform model training on the model training set using a random forest network algorithm to generate a trading decision training model; perform model optimization iteration on the trading decision training model using a model test set to generate a trading decision model;

[0059] Step S423: Import the first logistics transportation transaction data into the transaction decision model to make a transaction cost decision and generate transaction cost decision data.

[0060] The present invention divides the transaction adjustment rule data into data sets to obtain a model training set and a model test set. This step ensures the scientific nature of model training and testing, makes the model training process more effective, and can be verified on different data sets. The model training set is trained using a random forest network algorithm to generate a transaction decision training model. This step ensures that the model can efficiently learn and identify complex data patterns by using the advantages of the random forest algorithm, thereby improving the accuracy and reliability of the model. The transaction decision training model is optimized and iterated by the model test set to generate a transaction decision model. This process ensures the continuous optimization and improvement of the model, improves the generalization ability and prediction performance of the model, and ensures that the performance of the model in practical applications is more stable and accurate. The first logistics transportation transaction data is imported into the transaction decision model to make a transaction cost decision and generate transaction cost decision data. This step ensures the scientific nature and accuracy of the transaction cost decision by using the optimized model, thereby improving the rationality of the cost decision. Through the random forest algorithm and model optimization iteration, the accuracy and stability of the decision model can be effectively improved, thereby enhancing the reliability of the decision and reducing the errors and risks in the decision. Using advanced algorithms and optimization techniques to support data-driven decision-making makes transaction cost decisions more data- and model-based, avoiding the uncertainty caused by subjective judgment. By generating transaction decision models and decision data, it can help companies better manage the transaction process, optimize the cost structure, and improve overall transaction management efficiency. The model optimization and iteration process can improve the model's ability to adapt to different data scenarios, so that the model can still maintain a high degree of decision accuracy when facing a changing business environment.

[0061] The beneficial effect of the present invention is that by preprocessing and evaluating the comprehensive logistics data, standardized data can be generated, which helps to improve data quality and accuracy. By analyzing the cargo value assessment data, the best initial transportation mode can be formulated, which helps to improve transportation efficiency and reduce costs. By using sensors for intelligent perception and dynamic transportation mode selection, more reasonable transportation decisions can be made according to real-time conditions, improving the flexibility and adaptability of transportation. Through reverse logistics cost analysis and transaction fee adjustment, costs can be better controlled, and transaction fees can be adjusted when necessary to improve benefits. Constructing transaction adjustment rules and applying machine learning methods to make decisions on transaction fees can optimize the transaction process and improve the intelligence and efficiency of decision-making. Therefore, the present invention improves the accuracy and comprehensiveness of the construction of transaction decision models through standardized data processing, dynamic transportation mode selection, reverse logistics cost analysis and machine learning decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of the steps of a method for constructing a transaction decision model for logistics data analysis;

[0063] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0064] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0065] Figure 4 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0066] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0067] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0068] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0069] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0070] To achieve this, please refer to Figures 1 to 4 , a method for constructing a transaction decision model for logistics data analysis, the method comprising the following steps:

[0071] Step S1: Acquire comprehensive logistics data; perform data preprocessing on the comprehensive logistics data to generate standard comprehensive logistics data; perform cargo value assessment on the standard comprehensive logistics data to generate cargo value assessment data; perform initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate initial logistics transportation mode data;

[0072] Step S2: based on the cargo value assessment data, the first logistics transaction fee of the initial logistics transportation mode data is confirmed to obtain the first logistics transportation transaction data; the logistics transport cargo is intelligently sensed by the sensor to obtain the logistics cargo intelligent sensing data; the logistics cargo intelligent sensing data is dynamically selected to select the transportation mode to generate the dynamic transportation mode data; the transaction fee of the first logistics transportation transaction data is adjusted by the dynamic transportation mode data to generate the second logistics transportation transaction data;

[0073] Step S3: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data; reverse logistics cost analysis is performed on the cargo logistics delivery feedback data to generate reverse logistics cost data; transaction cost adjustment is performed on the first logistics transportation transaction data based on the reverse logistics cost data to generate third logistics transportation transaction data;

[0074] Step S4: construct transaction adjustment rules based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data; use machine learning methods to make transaction cost decisions on the first logistics and transportation transaction data based on the transaction adjustment rule data, thereby generating a transaction cost decision report.

[0075] The present invention eliminates noise and inconsistency in the data by preprocessing and standardizing the comprehensive logistics data, thereby ensuring the quality and consistency of the data. This makes subsequent analysis more accurate, reduces errors and deviations caused by data inconsistency, and thus improves the reliability of the overall decision. The value assessment of the goods ensures that the transportation decision can fully consider the economic importance of the goods. This enables high-value goods to be given priority when selecting transportation modes and allocating resources, optimizes resource allocation, and improves transportation efficiency and cost-effectiveness. Based on the initial transportation mode analysis of the goods value assessment data, a transportation plan that better meets the needs of the goods can be formulated. This provides a scientific basis for subsequent cost confirmation, helps to formulate a reasonable transportation plan and budget, and thus improves the accuracy of transportation decisions. By preliminarily confirming the logistics transaction costs, a preliminary basis is provided for budget formulation and cost control during the transportation process. This helps to avoid budget overruns, ensure that costs are controlled within an acceptable range, and optimize financial planning. By using sensor technology for intelligent perception, the status and environmental information of logistics goods can be obtained in real time, and the transportation mode can be dynamically adjusted. This flexibility improves transportation efficiency, can respond to changes in actual conditions in a timely manner, and reduce delays and losses. Adjusting initial shipping costs based on dynamic shipping mode data makes cost calculations more accurate and realistic. Adjusting costs based on real-time data can better reflect the actual costs during transportation and reduce problems caused by inaccurate cost estimates in static modes. Providing real-time feedback during transportation can promptly identify and resolve problems such as shipping delays and damage. This helps improve the quality of transportation services, increase customer satisfaction, and reduce additional costs caused by logistics problems. Reverse logistics cost analysis can comprehensively evaluate the costs of returns and reverse logistics, ensuring that all relevant costs are taken into account. This makes cost estimates more comprehensive and avoids budget shortfalls caused by ignoring reverse logistics costs. Adjusting costs based on reverse logistics cost data ensures that the final cost calculation is more accurate. By comprehensively considering all relevant costs, budget management is optimized and financial losses caused by inadequate consideration of reverse logistics costs are reduced. Building transaction adjustment rules can standardize the cost adjustment process and improve the consistency and repeatability of decisions. This reduces decision inconsistency caused by the lack of clear rules and ensures the transparency and fairness of the transaction cost adjustment process. Using machine learning methods to learn from transaction adjustment rule data, the decision process is optimized and the level of intelligence in decision making is improved. By analyzing historical data, an accurate cost decision report is generated, thereby enhancing the scientificity and accuracy of decision-making and further optimizing logistics cost management. Therefore, the present invention improves the accuracy and comprehensiveness of transaction decision model construction through standardized data processing, dynamic transportation mode selection, reverse logistics cost analysis and machine learning decision-making.

[0076] In the embodiment of the present invention, reference Figure 1The above is a schematic flow chart of the steps of a method for constructing a transaction decision model for logistics data analysis of the present invention. In this example, the method for constructing a transaction decision model for logistics data analysis includes the following steps:

[0077] Step S1: Acquire comprehensive logistics data; perform data preprocessing on the comprehensive logistics data to generate standard comprehensive logistics data; perform cargo value assessment on the standard comprehensive logistics data to generate cargo value assessment data; perform initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate initial logistics transportation mode data;

[0078] In an embodiment of the present invention, various data sources involved in logistics operations are identified, such as a transportation management system (TMS), a warehouse management system (WMS), a customer order system, etc. Comprehensive data related to logistics, including cargo information, transportation routes, timestamps, transportation costs, etc., are collected through an API interface, a data export function, or a real-time data stream. Missing values, outliers, and duplicate data are processed. For example, missing values ​​are filled, incorrect data formats are corrected, etc. Data in different formats are uniformly converted into a standard format to ensure data consistency. For example, the timestamp format is unified into the ISO 8601 format. Numerical data is normalized or standardized to ensure that data is analyzed at the same scale. Evaluation indicators of goods, such as market value, transportation cost, and importance of goods, are determined according to business needs. The value of each piece of goods is evaluated using data analysis techniques, such as statistical analysis, weighted scoring models, etc. For example, the total value is calculated by the market price and transportation cost of the goods. The results of the goods value evaluation are organized into structured data, including the value score of each piece of goods, the basis for evaluation, etc. According to the goods value evaluation data, the dimensions of the initial transportation mode analysis are defined, such as the mode of transportation, the transportation route, and the timeliness requirements. Select specific analysis models, such as cluster analysis, decision tree models, etc., to evaluate the transportation mode of goods. Analyze the logistics transportation cargo information data to generate initial transportation mode data. The analysis results should include recommended transportation methods, optimized transportation routes, etc. Organize the analysis results into a clear report or data set, including recommended transportation modes, expected transportation time and cost, etc. Use visualization tools (such as charts, dashboards, etc.) to display the initial transportation mode data so that relevant personnel can understand and make decisions.

[0079] Step S2: based on the cargo value assessment data, the first logistics transaction fee of the initial logistics transportation mode data is confirmed to obtain the first logistics transportation transaction data; the logistics transport cargo is intelligently sensed by the sensor to obtain the logistics cargo intelligent sensing data; the logistics cargo intelligent sensing data is dynamically selected to select the transportation mode to generate the dynamic transportation mode data; the transaction fee of the first logistics transportation transaction data is adjusted by the dynamic transportation mode data to generate the second logistics transportation transaction data;

[0080] In the embodiment of the present invention, the main factors affecting the logistics transaction costs, such as transportation distance, transportation mode, cargo weight, insurance cost, etc., are determined. The cargo value assessment data is matched with the initial transportation mode data to identify the cost components under each transportation mode. The logistics transaction costs under the initial transportation mode are calculated using a predetermined cost model, including freight calculation formula, additional costs, etc. The calculation results are integrated to generate a data set containing the cost information of each logistics transaction, which is marked as "first logistics transportation transaction data". Sensors are installed on the goods and transportation tools to monitor key parameters in the transportation process in real time, such as temperature, humidity, vibration, etc. The sensors transmit the collected data to the central system in real time through a wireless network or a wired network. The collected sensor data is stored and preliminarily processed to form "logistics cargo intelligent perception data". The intelligent perception data is processed using data analysis tools and algorithms to identify the actual conditions in the transportation process (such as whether the goods are under optimal environmental conditions). Based on the analysis results, the transportation mode is dynamically adjusted, such as selecting a different transportation mode or adjusting the transportation route to optimize the transportation conditions. The analysis results and adjustment suggestions are integrated to generate dynamic transportation mode data, including recommended transportation modes and adjustment measures. Define a cost adjustment model based on dynamic transportation patterns, such as adjusting costs caused by changes in the transportation environment. Calculate the cost adjustment amount for each transaction based on the dynamic transportation pattern data and the cost adjustment model. Combine the adjusted cost information with the original first logistics transportation transaction data to generate new "second logistics transportation transaction data", including the updated cost information and adjustment basis.

[0081] Step S3: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data; reverse logistics cost analysis is performed on the cargo logistics delivery feedback data to generate reverse logistics cost data; transaction cost adjustment is performed on the first logistics transportation transaction data based on the reverse logistics cost data to generate third logistics transportation transaction data;

[0082] In an embodiment of the present invention, real-time data during the transportation of goods, such as delivery time, transportation status, damage to goods, etc., are collected by using tools such as transportation management systems and real-time tracking systems. The specific conditions of the delivery of goods, including whether the delivery is on time and whether the goods are complete, are recorded through customer feedback, transportation personnel reports or automated systems. The collected feedback data is organized into structured "goods logistics delivery feedback data", including delivery status, customer satisfaction, abnormal conditions, etc. Factors affecting reverse logistics costs, such as return processing fees, transportation return fees, and goods reprocessing fees, are determined. Reverse logistics costs are calculated using data analysis tools and models. This includes identifying the reasons for returns from feedback data and evaluating return processing and transportation costs. The analysis results are integrated to generate "reverse logistics cost data", which includes cost information and its components for each return or problem handling. A cost adjustment model is developed based on the reverse logistics cost data. The model should consider the impact of return costs, processing fees, etc. on transaction costs. The adjustment model is applied to calculate the cost adjustment amount for each transaction based on the reverse logistics cost data. The adjusted cost information is combined with the original first logistics and transportation transaction data to generate "third logistics and transportation transaction data", including the updated cost information and adjustment basis.

[0083] Step S4: construct transaction adjustment rules based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data; use machine learning methods to make transaction cost decisions on the first logistics and transportation transaction data based on the transaction adjustment rule data, thereby generating a transaction cost decision report.

[0084] In an embodiment of the present invention, by integrating the third logistics and transportation transaction data and the second logistics and transportation transaction data, common features and adjustment requirements are identified and extracted. The framework of the transaction adjustment rules is defined, including the basic composition of the rules, adjustment conditions, adjustment factors, etc. Common adjustment patterns and conditions are identified using data mining techniques (such as association rule mining). Factors (such as transportation delays, cost changes, and reverse logistics costs) that affect transaction costs are analyzed. The mined adjustment patterns are converted into specific adjustment rules, and a set of rules including adjustment conditions, adjustment strategies, and processing procedures are formulated. The rules are organized into a structured data set, called "transaction adjustment rule data", for subsequent machine learning applications. Missing values, outliers, etc. are processed to ensure data quality, and features related to cost decisions, such as transportation time, cargo type, reverse logistics costs, etc., are extracted from the first logistics and transportation transaction data. Specific machine learning algorithms are selected according to data characteristics and business needs, such as regression analysis, decision trees, random forests, support vector machines, etc. The machine learning model is trained using the transaction adjustment rule data and the first logistics and transportation transaction data to learn how to predict transaction costs based on the adjustment rules. Evaluate model performance through cross-validation and other techniques to ensure its accuracy in transaction cost decisions, adjust model parameters based on validation results, and optimize model performance. Apply the trained model to the First Logistics transportation transaction data to make transaction cost predictions and decisions. Organize the prediction results and decision basis to generate a "transaction cost decision report". The report should include a detailed analysis of the decision, adjustment suggestions, and the final cost prediction results.

[0085] Preferably, step S1 comprises the following steps:

[0086] Step S11: Obtaining comprehensive logistics data;

[0087] Step S12: preprocessing the logistics comprehensive data to generate standard logistics comprehensive data, wherein the data preprocessing includes data cleaning, data missing value filling and data standardization;

[0088] Step S13: screening the transported goods data from the standard logistics comprehensive data to obtain logistics transported goods information data;

[0089] Step S14: perform cargo value assessment on the logistics transportation cargo information data to generate cargo value assessment data; perform initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate logistics initial transportation mode data, wherein the logistics initial transportation mode data includes air transportation mode data, sea transportation mode data and land transportation mode data.

[0090] In an embodiment of the present invention, comprehensive logistics data is collected from different data sources (such as logistics management systems, transportation company databases, sensor data, order systems, etc.). The collected data types include transportation routes, cargo types, weights, volumes, transportation time, costs, transportation conditions (such as temperature, humidity, etc.). Detect and delete duplicate records, correct erroneous values ​​in the data (such as inconsistent formats, outliers, etc.), and ensure that the formats of all data fields are consistent (such as date formats, units, etc.). Identify the location of missing values ​​in the data, and select appropriate filling methods (such as mean filling, median filling, interpolation, etc.) according to the nature of the missing data. Standardize the data (such as z-score standardization) or normalize (such as scaling the data to a range of 0-1) to ensure that the data meets unified standards and scales for subsequent processing. Determine screening criteria, such as: cargo categories (such as perishables, valuables, etc.), transportation requirements (such as urgent items, ordinary items, etc.), and transportation areas or route restrictions. Use data query and filtering tools (such as SQL queries, data analysis software, etc.) to filter the data, extract qualified cargo information data, and form logistics transportation cargo information data. Analyze the type, value, weight, volume and other characteristics of the goods, and consider the impact of market demand and trends on the value of the goods. Collect market price data for similar goods and analyze market price trends, including historical price trends and future forecasts. Calculate the production cost and procurement cost of the goods, and conduct a comprehensive market value assessment of the goods in combination with market trends and goods characteristics. Use assessment models or tools (such as regression analysis, valuation models, etc.) to generate goods value assessment data. Select appropriate transportation modes based on the goods value assessment data and goods characteristics, and generate data for different transportation modes, including relevant information on air, sea and land transportation, including information on the cost, time requirements, applicable conditions, etc. of each transportation mode.

[0091] Preferably, step S14 comprises the following steps:

[0092] Step S141: Analyze the characteristics of the goods transported by the logistics transport information data to generate goods characteristic data;

[0093] Step S142: performing market price trend analysis of similar items on the logistics transport cargo information data based on the cargo characteristic data to generate cargo market trend data;

[0094] Step S143: Performing a comprehensive value assessment of the goods through the goods market trend data and the goods characteristic data to generate goods value assessment data, wherein the comprehensive value assessment of the goods includes cost estimation and market value analysis;

[0095] Step S144: Performing a transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate logistics transportation mode data.

[0096] In an embodiment of the present invention, by collecting basic information of goods, such as type, weight, volume, packaging method, storage requirements, etc., obtain special transportation requirements of goods, such as temperature control, humidity control, vibration sensitivity, etc. Classify the characteristics of goods, such as perishability, packaging type, transportation environment requirements, etc. Use statistical methods to analyze the distribution and frequency of goods characteristic data, generate goods characteristic data, and generate a detailed goods characteristic report, including statistical data and analysis results of various characteristics. According to the goods characteristic data, identify similar items on the market. Similarity analysis algorithms (such as K-nearest neighbor algorithms) can be used to find items with similar characteristics. Collect price data and trend information of similar items from market databases, industry reports, online platforms, etc. Use time series analysis, regression analysis, etc. to analyze the market price trends of similar items, and form a detailed data report on market price fluctuations and trends. Comprehensive estimation is performed based on the production cost, procurement cost, transportation cost, etc. of the goods. Cost plus method or other cost estimation methods can be used. Apply market price trend data to cargo characteristic data to conduct a comprehensive assessment of market value. Use valuation models (such as weighted average method, comparative market analysis, etc.) to calculate the market value of cargo, generate cargo value assessment data reports including cost estimation and market value analysis, and describe the comprehensive value of cargo in detail. Based on the cargo value assessment data, define the criteria for selecting a transportation mode, such as cost-effectiveness ratio, transportation time, cargo safety, etc. Evaluate different transportation modes (such as air, sea, and land transportation), including the advantages and disadvantages of each mode, applicable scenarios, etc. Select the most appropriate transportation mode based on the comprehensive value of the cargo, characteristic requirements, and market price trends, generate data on different transportation modes, including estimated costs, transportation time, applicable conditions, etc., generate a detailed logistics transportation mode analysis report, and provide selection recommendations and analysis basis.

[0097] Preferably, step S144 includes the following steps:

[0098] Step S1441: Locate the starting point and the end point of the transportation of the goods in the logistics transportation, and obtain the starting point data and the end point data of the transportation of the goods;

[0099] Step S1442: performing cargo characteristic demand analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate cargo characteristic demand data;

[0100] Step S1443: Evaluate the transportation mode of the cargo transportation starting point data and the cargo transportation destination data according to the cargo characteristic demand data to generate logistics initial transportation mode data, wherein the logistics initial transportation mode data includes air transportation mode data, sea transportation mode data and land transportation mode data.

[0101] In an embodiment of the present invention, the transportation starting point and end point data of the goods are extracted from the logistics system or order management system. These data include warehouse locations, distribution centers, customer addresses, etc. The starting point and end point addresses are converted into geographic coordinates (longitude and latitude). Geocoding services (such as Google Maps API, OpenStreetMap, etc.) can be used to complete this task, generating a data set containing the coordinates of the starting point and the end point of the goods transportation. To ensure the accuracy of the starting point and end point coordinates, it can be verified by comparing with the actual address or using a map visualization tool, where the transportation starting point data contains the geographic coordinates of all starting points, and the transportation end point data contains the geographic coordinates of all end points. According to the goods value assessment data, the characteristic requirements of the goods (such as temperature control, humidity control, vibration protection, etc.) are identified. For example, high-value goods require more stringent transportation conditions, and the characteristics of the goods are classified, such as perishable goods, valuables, etc., and the characteristics of the goods are generated. The requirements include transportation conditions and requirements for each type of goods, and a goods characteristic requirement analysis report is written, which describes the characteristics and requirements of each type of goods in detail. According to the characteristics of the goods, the evaluation criteria are formulated. These criteria include transportation cost, transportation time, safety, applicable transportation mode (such as air, sea, land), etc. Analyze the adaptability of each transportation mode to different cargo characteristics. For example, high-value and perishable goods are more suitable for air transportation, while large-volume goods with less urgent transportation time can choose sea transportation. Match the transportation mode with the cargo characteristics, select the most appropriate transportation mode, and generate initial transportation mode data.

[0102] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0103] Step S21: Perform logistics transportation path planning based on the logistics initial transportation mode data to generate logistics transportation path planning data; use the goods value assessment data to confirm the first logistics transaction fee of the logistics transportation path planning data to obtain the first logistics transportation transaction data;

[0104] Step S22: Using sensors to perform intelligent perception of logistics transport goods to obtain intelligent perception data of logistics goods; building a sensor network for the intelligent perception data of logistics goods to obtain a transport goods sensor network; collecting real-time cargo transportation information on the transport goods sensor network to obtain real-time cargo transportation data;

[0105] Step S23: performing edge computing on the real-time data of cargo transportation to generate cargo transportation edge computing data; performing dynamic transportation mode selection on the cargo transportation edge computing data to generate dynamic transportation mode data, wherein the dynamic transportation mode data includes a single transportation mode and a multi-mode transportation mode;

[0106] Step S24: When it is confirmed that the dynamic transportation mode data is a multi-modal transportation mode, the transaction fee of the first logistics transportation transaction data is adjusted based on the air transportation mode data, the sea transportation mode data and the land transportation mode data to generate the second logistics transportation transaction data; when it is confirmed that the dynamic transportation mode data is a single transportation mode, the first logistics transportation transaction data is maintained unchanged.

[0107] In the embodiment of the present invention, by selecting a specific path planning algorithm (such as Dijkstra algorithm, genetic algorithm, etc.), the path is optimized according to the transportation mode and demand. Based on the transportation mode data and the starting and ending point information, the optimal logistics transportation path is calculated. Factors to be considered include cost, time, feasibility of the transportation mode, etc. The optimal path under each transportation mode and its detailed information (such as passing places, estimated time, transportation cost, etc.) are included. According to the cargo value assessment data and the logistics transportation path planning data, the transportation cost of each path is calculated, including the transaction cost confirmation data of the first round, and the cost structure under the path planning is described. According to the requirements of the cargo characteristics, specific sensors (such as temperature sensors, humidity sensors, vibration sensors, etc.) are selected. Sensors are installed during the transportation of goods to ensure that the required data can be collected in real time. Real-time environmental data of logistics goods, including temperature, humidity, vibration, etc., are collected through sensors. The sensor data is connected to the central data system through the network to form a sensor network. The sensor network is monitored to collect real-time data during the transportation of goods, including all real-time monitoring data obtained from the sensor network. The real-time data is transmitted to the edge computing node for data processing. The edge computing nodes are located close to the data source to reduce latency and bandwidth consumption, and analyze real-time data, including real-time calculation of data such as temperature, humidity, and vibration. The edge computing results generated based on the real-time data analysis include real-time status evaluation and anomaly detection results. Based on the edge computing data, the most suitable transportation mode is determined. A single transportation mode or a multi-modal transportation mode is dynamically selected, and dynamic transportation mode data is generated to describe the selected transportation mode and its applicability. Based on the dynamic transportation mode data, it is confirmed whether it is a single transportation mode or a multi-modal transportation mode. Based on the air transportation mode data, the sea transportation mode data, and the land transportation mode data, the first logistics transportation transaction data is adjusted. Considering the cost and benefit of each mode, the second logistics transportation transaction data is generated, including the adjusted cost structure and details. If the dynamic transportation mode data is a single transportation mode, the first logistics transportation transaction data is maintained unchanged.

[0108] Preferably, performing dynamic transportation mode selection on the cargo transportation edge computing data to generate dynamic transportation mode data includes:

[0109] Perform environmental data analysis on the edge computing data of cargo transportation to generate cargo transportation environmental data, where the environmental data analysis includes temperature analysis, humidity analysis, and vibration analysis;

[0110] Cargo status evaluation is performed on cargo transportation environment data to generate cargo status evaluation data; conditional judgment is performed on cargo status evaluation data based on preset transportation conditions, and when the cargo status evaluation data does not meet the preset transportation conditions, transportation mode switching is performed based on the cargo status evaluation data to generate dynamic transportation mode data;

[0111] Among them, the transportation mode switching based on cargo status assessment data includes:

[0112] Perform environmental anomaly identification on the cargo transportation environmental data to obtain abnormal cargo transportation environmental data; when the temperature in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the sea transportation mode data; when the humidity in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the land transportation mode data; when the vibration in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the sea transportation mode data.

[0113] In an embodiment of the present invention, real-time cargo transportation data including temperature, humidity and vibration data are collected from edge computing nodes. The temperature change during cargo transportation is evaluated to detect whether there is a temperature fluctuation beyond a preset range. The humidity data is analyzed to check whether the humidity meets the transportation requirements and to identify humidity abnormalities. The vibration data is monitored to analyze the vibration intensity and frequency to determine whether there is a vibration abnormality that causes damage to the cargo. The temperature, humidity and vibration analysis results are integrated into cargo transportation environment data to describe the current environmental state. Based on the environmental data, the state of the cargo is evaluated, including whether it is affected by environmental factors, such as too high temperature, too high humidity or too strong vibration, to form cargo state evaluation data, including whether the cargo meets the preset transportation conditions. Define the environmental conditions (such as temperature range, humidity range and vibration range) that the cargo must meet during transportation. Compare the cargo state evaluation data with the preset conditions to determine whether the cargo meets the transportation requirements. If the cargo state evaluation data does not meet the preset transportation conditions, the transportation mode is switched. Identify temperature abnormalities to determine whether it is necessary to switch from air transportation mode to sea transportation mode or other suitable transportation modes. Identify humidity abnormalities to determine whether it is necessary to switch from air transportation mode to land transportation mode or other suitable transportation modes. Identify abnormal vibration conditions and decide whether to switch from air transport mode to sea transport mode or other suitable transport modes. Generate dynamic transport mode data based on the results of environmental anomaly identification and condition judgment to describe the currently selected transport mode (such as air transport, sea transport, land transport or multi-modal transport mode). Apply the switched transport mode to the actual logistics process and adjust the transport arrangements to ensure the safety of the goods and meet the transport conditions.

[0114] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0115] Step S31: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data;

[0116] Step S32: evaluating the feedback satisfaction of the cargo logistics delivery feedback data to generate cargo logistics delivery satisfaction data;

[0117] Step S33: performing reverse logistics cost analysis on the cargo logistics delivery feedback data through the cargo logistics delivery satisfaction data to generate reverse logistics cost data;

[0118] Step S34: adjusting the transaction costs of the first logistics and transportation transaction data based on the reverse logistics cost data to generate third logistics and transportation transaction data.

[0119] In the embodiment of the present invention, real-time data of cargo transportation is collected, including delivery status, transportation timeliness, cargo status, etc. Delivery feedback is obtained from the consignee, including the integrity, timeliness, service quality, etc. of the cargo. The feedback information is sorted and summarized to form cargo logistics delivery feedback data. The feedback data should include indicators such as delivery time, cargo status, and customer satisfaction. The cargo logistics delivery feedback data is analyzed to calculate the customer satisfaction score. The cargo logistics delivery satisfaction data is generated through methods such as satisfaction surveys and scoring systems. These data should include the customer's overall satisfaction with the delivery process, specific problem feedback and suggestions, etc. According to the satisfaction data, the feedback is divided into positive feedback and negative feedback. The return situation in the negative feedback is counted, the return rate and frequency are calculated, and the return frequency analysis data is generated. Based on the satisfaction data and the return frequency data, the reverse logistics cost elements are identified. The elements considered include return processing fees, transportation costs, repackaging costs, etc. Comprehensively analyze the various cost elements, and generate reverse logistics cost data. These data should include total cost, cost structure and trend analysis. Integrate the reverse logistics cost data with the first logistics and transportation transaction data to identify the expense items that need to be adjusted. Based on the reverse logistics cost analysis results, adjust the transaction costs of the first logistics and transportation transaction data, including return costs, compensation costs, etc. Update the transaction records and generate adjusted third-party logistics and transportation transaction data to ensure that the costs accurately reflect the actual costs.

[0120] Preferably, step S33 includes the following steps:

[0121] Step S331: performing feedback score matching on the cargo logistics delivery satisfaction data and the cargo logistics delivery feedback data to generate feedback score matching data; performing correlation analysis on the feedback score matching data to generate delivery satisfaction influencing factor data;

[0122] Step S332: classify the goods feedback on the goods logistics delivery feedback data to generate positive feedback goods data and negative feedback goods data; perform return frequency analysis on the negative feedback goods data to generate return frequency analysis data;

[0123] Step S333: Based on the delivery satisfaction influencing factor data, the return frequency analysis data and the positive feedback goods data are used to identify the reverse logistics cost elements, and generate reverse logistics cost element data;

[0124] Step S334: Perform reverse logistics cost analysis on the goods logistics delivery feedback data through reverse logistics cost element data to generate reverse logistics cost data, wherein the reverse logistics cost analysis includes total cost summary, cost structure analysis and trend analysis.

[0125] In an embodiment of the present invention, the satisfaction data of cargo logistics delivery and the feedback data of cargo logistics delivery are scored and matched. Generally, the evaluation level (such as satisfaction score, service quality evaluation, etc.) in the feedback data is compared with the score in the satisfaction data to generate feedback score matching data. Correlation analysis is performed on the feedback score matching data to determine which factors have a significant impact on satisfaction. Statistical methods (such as Pearson correlation coefficient, regression analysis) are used to identify the main influencing factors. Delivery satisfaction influencing factor data are generated, and these factors include transportation timeliness, cargo status, service quality, etc. The cargo logistics delivery feedback data is classified into positive feedback and negative feedback. Positive feedback refers to customer satisfaction, good service, etc., while negative feedback involves returns, complaints and other issues. A detailed return frequency analysis is performed on the negative feedback cargo data. The frequency of occurrence of different types of returns (such as damage, misdelivery, etc.) is counted, and return frequency analysis data is generated. These data help understand the scale and nature of the return problem. Based on the delivery satisfaction influencing factor data, the return frequency analysis data and the positive feedback cargo data are used to identify the reverse logistics cost elements. Identify the main cost elements, including return processing fees, transportation fees, repackaging fees, processing personnel fees, etc., and organize the identified cost elements into reverse logistics cost element data. These data help to understand the composition and impact of various costs in the reverse logistics process. Summarize all reverse logistics cost element data and calculate the total cost. Including return processing fees, transportation fees, etc., to obtain the total cost of reverse logistics. Analyze the structure of the total cost and determine the proportion of each cost element in the total cost. For example, the proportion of return processing fees in the total cost, etc. Perform time trend analysis on reverse logistics cost data to identify cost change trends. For example, whether the cost increases over time, whether there are seasonal fluctuations, etc. Summarize the analysis results and generate reverse logistics cost data. These data will include total cost, cost structure and trend analysis results, providing a basis for subsequent cost adjustment and optimization.

[0126] As an example of the present invention, refer to Figure 4 As shown, in this example, step S4 includes:

[0127] Step S41: constructing a transaction adjustment rule based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data;

[0128] Step S42: constructing a transaction decision model through the transaction adjustment rule data to obtain a transaction decision model; using the transaction decision model to make a transaction cost decision on the first logistics transportation transaction data, thereby generating transaction cost decision data;

[0129] Step S43: Visualize the transaction fee decision data to generate a transaction fee decision report.

[0130] In the embodiment of the present invention, the third logistics transportation transaction data and the second logistics transportation transaction data are obtained. These data should include information such as the cost, mode, time, and cargo characteristics of each transportation transaction. The collected data is preprocessed, including data cleaning, missing value processing, data standardization, etc., to ensure data quality. The key influencing factors in the third logistics transportation transaction data and the second logistics transportation transaction data, such as transportation mode, cargo characteristics, route, time, etc., are analyzed. Based on the influencing factors, transaction adjustment rules are constructed. For example, according to different transportation modes, the cost calculation method is adjusted; according to different cargo characteristics, different cost adjustment rules are set. The constructed rules are encoded into a logical algorithm or a decision tree to generate transaction adjustment rule data. The rules include price adjustment formulas, priority rules, conditional judgments, etc. The validity of the transaction adjustment rules is tested using historical data to ensure that the rules can accurately reflect the actual situation. According to the verification results, the transaction adjustment rules are optimized to improve the accuracy and adaptability of the rules. The constructed transaction adjustment rule data is applied to model construction. The data set usually includes a training set and a test set for training and verifying the transaction decision model. Select a suitable machine learning or optimization algorithm (such as random forest, support vector machine, regression analysis, etc.) to construct the transaction decision model. The selection criteria include data characteristics, prediction targets, model complexity, etc. Use the training set data to train the selected algorithm and generate a preliminary transaction decision model. Use the test set data to evaluate the model performance and test the model's performance on different data. Adjust and optimize the model parameters based on the evaluation results to improve the model's prediction accuracy and stability. Input the first logistics transportation transaction data into the optimized transaction decision model to make transaction fee decisions. Generate transaction fee decision data based on the model output, which includes adjusted fee information. Select suitable data visualization tools or platforms, such as Tableau, Power BI, Matplotlib, etc. Convert the transaction fee decision data into various charts, such as bar charts, pie charts, line charts, etc., for easy analysis and presentation. The charts should clearly show the differences, trends, and patterns before and after the fee adjustment. Design an interactive visualization interface that allows users to filter, sort, and deeply analyze the data. Based on the data visualization results, write a detailed transaction fee decision report, which should include data analysis results, chart presentation, decision basis, and suggestions. Review and proofread the report content to ensure accuracy and completeness, and publish the final report to relevant stakeholders, including management, finance department, customers, etc.

[0131] Preferably, step S42 includes the following steps:

[0132] Step S421: dividing the transaction adjustment rule data into data sets to obtain a model training set and a model test set;

[0133] Step S422: Perform model training on the model training set using a random forest network algorithm to generate a trading decision training model; perform model optimization iteration on the trading decision training model using a model test set to generate a trading decision model;

[0134] Step S423: Import the first logistics transportation transaction data into the transaction decision model to make a transaction cost decision and generate transaction cost decision data.

[0135] In an embodiment of the present invention, by ensuring that a complete data set containing transaction adjustment rule data has been obtained. The data set should include various types of data related to logistics and transportation, such as cost information, transportation mode, cargo characteristics, etc. Clean the data to remove missing values, outliers and duplicate data to ensure data quality. Standardize the data, such as normalization or standardization, to make the data suitable for model training. Divide the data according to common proportions (such as 80% training set and 20% test set). This can be achieved using Python's train_test_split function (from sklearn.model_selection). The training set is used to train the transaction decision model and contains 80% of the data, and the test set is used to evaluate and optimize the transaction decision model and contains 20% of the data. Select the random forest algorithm for model training. Random forest is an ensemble learning method that improves the accuracy and stability of the model by building multiple decision trees. Use RandomForestRegressor (for regression problems) or RandomForestClassifier (for classification problems) in Python to select a suitable model according to task requirements. Call the fit method of the model and train it using the training set data. The sample code is as follows: Use the test set data to evaluate the trained model. Common evaluation indicators include mean square error (MSE), mean absolute error (MAE), etc. Adjust the model's hyperparameters, such as the number of trees, maximum depth, etc., based on the evaluation results to improve model performance. You can use GridSearchCV or RandomizedSearchCV for hyperparameter tuning. Use cross-validation technology to further optimize model performance and reduce overfitting. Import the first logistics transportation transaction data into the optimized transaction decision model. Ensure that the data format is consistent with the model input requirements. Use the model's predict method to predict the first logistics transportation transaction data, generate transaction cost decision data, and organize the generated transaction cost decision data into a format suitable for reporting and analysis, such as a data table or database record.

[0136] The beneficial effect of the present invention is that by preprocessing and evaluating the comprehensive logistics data, standardized data can be generated, which helps to improve data quality and accuracy. By analyzing the cargo value assessment data, the best initial transportation mode can be formulated, which helps to improve transportation efficiency and reduce costs. By using sensors for intelligent perception and dynamic transportation mode selection, more reasonable transportation decisions can be made according to real-time conditions, improving the flexibility and adaptability of transportation. Through reverse logistics cost analysis and transaction fee adjustment, costs can be better controlled, and transaction fees can be adjusted when necessary to improve benefits. Constructing transaction adjustment rules and applying machine learning methods to make decisions on transaction fees can optimize the transaction process and improve the intelligence and efficiency of decision-making. Therefore, the present invention improves the accuracy and comprehensiveness of the construction of transaction decision models through standardized data processing, dynamic transportation mode selection, reverse logistics cost analysis and machine learning decision-making.

[0137] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0138] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for constructing a transaction decision model for logistics data analysis, characterized in that: The following steps are involved: Step S1: Obtain comprehensive logistics data; Preprocess the comprehensive logistics data to generate standard comprehensive logistics data; Cargo value assessment is performed on standard logistics comprehensive data to generate cargo value assessment data; initial transportation mode analysis is performed on logistics transportation cargo information data based on cargo value assessment data to generate initial logistics transportation mode data; Step S2: based on the cargo value assessment data, the first logistics transaction fee of the initial logistics transportation mode data is confirmed to obtain the first logistics transportation transaction data; the logistics transport cargo is intelligently sensed by the sensor to obtain the logistics cargo intelligent sensing data; the dynamic transportation mode is selected for the logistics cargo intelligent sensing data to generate the dynamic transportation mode data; Adjusting the transaction fee of the first logistics and transportation transaction data by using the dynamic transportation mode data to generate second logistics and transportation transaction data; Step S2 includes the following steps: Step S21: Perform logistics transportation path planning based on the logistics initial transportation mode data to generate logistics transportation path planning data; use the goods value assessment data to confirm the first logistics transaction fee of the logistics transportation path planning data to obtain the first logistics transportation transaction data; Step S22: Using sensors to perform intelligent perception of logistics transport goods to obtain intelligent perception data of logistics goods; building a sensor network for the intelligent perception data of logistics goods to obtain a transport goods sensor network; collecting real-time cargo transportation information on the transport goods sensor network to obtain real-time cargo transportation data; Step S23: performing edge computing on the real-time data of cargo transportation to generate cargo transportation edge computing data; performing dynamic transportation mode selection on the cargo transportation edge computing data to generate dynamic transportation mode data, wherein the dynamic transportation mode data includes a single transportation mode and a multi-mode transportation mode; Step S24: when it is confirmed that the dynamic transportation mode data is a multi-mode transportation mode, the transaction fee of the first logistics transportation transaction data is adjusted based on the air transportation mode data, the sea transportation mode data and the land transportation mode data to generate second logistics transportation transaction data; when it is confirmed that the dynamic transportation mode data is a single transportation mode, the first logistics transportation transaction data is maintained unchanged; Step S3: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data; reverse logistics cost analysis is performed on the cargo logistics delivery feedback data to generate reverse logistics cost data; transaction cost adjustment is performed on the first logistics transportation transaction data based on the reverse logistics cost data to generate third logistics transportation transaction data; Step S4: construct transaction adjustment rules based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data; use machine learning methods to make transaction cost decisions on the first logistics and transportation transaction data based on the transaction adjustment rule data, thereby generating a transaction cost decision report.

2. The method for constructing a transaction decision model for logistics data analysis according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtaining comprehensive logistics data; Step S12: preprocessing the logistics comprehensive data to generate standard logistics comprehensive data, wherein the data preprocessing includes data cleaning, data missing value filling and data standardization; Step S13: screening the transported goods data from the standard logistics comprehensive data to obtain logistics transported goods information data; Step S14: perform cargo value assessment on the logistics transportation cargo information data to generate cargo value assessment data; perform initial transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate logistics initial transportation mode data, wherein the logistics initial transportation mode data includes air transportation mode data, sea transportation mode data and land transportation mode data.

3. The method for constructing a transaction decision model for logistics data analysis according to claim 2, characterized in that: Step S14 includes the following steps: Step S141: Analyze the characteristics of the goods transported by the logistics transport information data to generate goods characteristic data; Step S142: performing market price trend analysis of similar items on the logistics transport cargo information data based on the cargo characteristic data to generate cargo market trend data; Step S143: Performing a comprehensive value assessment of the goods through the goods market trend data and the goods characteristic data to generate goods value assessment data, wherein the comprehensive value assessment of the goods includes cost estimation and market value analysis; Step S144: Performing a transportation mode analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate logistics transportation mode data.

4. The method for constructing a transaction decision model for logistics data analysis according to claim 3 is characterized in that: Step S144 includes the following steps: Step S1441: Locate the starting point and the end point of the transportation of the goods in the logistics transportation, and obtain the starting point data and the end point data of the transportation of the goods; Step S1442: performing cargo characteristic demand analysis on the logistics transportation cargo information data based on the cargo value assessment data to generate cargo characteristic demand data; Step S1443: Evaluate the transportation mode of the cargo transportation starting point data and the cargo transportation destination data according to the cargo characteristic demand data to generate logistics initial transportation mode data, wherein the logistics initial transportation mode data includes air transportation mode data, sea transportation mode data and land transportation mode data.

5. The method for constructing a transaction decision model for logistics data analysis according to claim 1, characterized in that: Dynamically select the transportation mode for the edge computing data of cargo transportation, and generate dynamic transportation mode data including: Perform environmental data analysis on the edge computing data of cargo transportation to generate cargo transportation environmental data, where the environmental data analysis includes temperature analysis, humidity analysis, and vibration analysis; Cargo status evaluation is performed on cargo transportation environment data to generate cargo status evaluation data; conditional judgment is performed on cargo status evaluation data based on preset transportation conditions, and when the cargo status evaluation data does not meet the preset transportation conditions, transportation mode switching is performed based on the cargo status evaluation data to generate dynamic transportation mode data; Among them, the transportation mode switching based on cargo status assessment data includes: Perform environmental anomaly identification on the cargo transportation environmental data to obtain abnormal cargo transportation environmental data; when the temperature in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the sea transportation mode data; when the humidity in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the land transportation mode data; when the vibration in the abnormal cargo transportation environmental data is confirmed to be abnormal, switch between the air transportation mode data and the sea transportation mode data.

6. The method for constructing a transaction decision model for logistics data analysis according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Cargo transportation delivery feedback is performed according to the real-time cargo transportation data to obtain cargo logistics delivery feedback data; Step S32: evaluating the feedback satisfaction of the cargo logistics delivery feedback data to generate cargo logistics delivery satisfaction data; Step S33: performing reverse logistics cost analysis on the cargo logistics delivery feedback data through the cargo logistics delivery satisfaction data to generate reverse logistics cost data; Step S34: adjusting the transaction costs of the first logistics and transportation transaction data based on the reverse logistics cost data to generate third logistics and transportation transaction data.

7. The method for constructing a transaction decision model for logistics data analysis according to claim 6, characterized in that: Step S33 includes the following steps: Step S331: performing feedback score matching on the cargo logistics delivery satisfaction data and the cargo logistics delivery feedback data to generate feedback score matching data; performing correlation analysis on the feedback score matching data to generate delivery satisfaction influencing factor data; Step S332: classify the goods feedback on the goods logistics delivery feedback data to generate positive feedback goods data and negative feedback goods data; perform return frequency analysis on the negative feedback goods data to generate return frequency analysis data; Step S333: Based on the delivery satisfaction influencing factor data, the return frequency analysis data and the positive feedback goods data are used to identify the reverse logistics cost elements, and generate reverse logistics cost element data; Step S334: Perform reverse logistics cost analysis on the goods logistics delivery feedback data through reverse logistics cost element data to generate reverse logistics cost data, wherein the reverse logistics cost analysis includes total cost summary, cost structure analysis and trend analysis.

8. The method for constructing a transaction decision model for logistics data analysis according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: constructing a transaction adjustment rule based on the third logistics and transportation transaction data and the second logistics and transportation transaction data to generate transaction adjustment rule data; Step S42: constructing a transaction decision model through the transaction adjustment rule data to obtain a transaction decision model; using the transaction decision model to make a transaction cost decision on the first logistics transportation transaction data, thereby generating transaction cost decision data; Step S43: Visualize the transaction fee decision data to generate a transaction fee decision report.

9. The method for constructing a transaction decision model for logistics data analysis according to claim 1, characterized in that: Step S42 includes the following steps: Step S421: dividing the transaction adjustment rule data into data sets to obtain a model training set and a model test set; Step S422: Perform model training on the model training set using a random forest network algorithm to generate a trading decision training model; perform model optimization iteration on the trading decision training model using a model test set to generate a trading decision model; Step S423: Import the first logistics transportation transaction data into the transaction decision model to make a transaction cost decision and generate transaction cost decision data.

Images