Big data driven logistics track intelligent prediction and path optimization system and method

Abstract

The invention provides a big-data-driven logistics track intelligent prediction and path optimization system and method, and relates to the technical field of logistics transportation management, and the method comprises the steps: collecting historical logistics track data which comprises transportation logistics track data, real-time traffic information and weather condition data; performing preprocessing operation on the historical logistics track data to generate corresponding target logistics track data; training and verifying the target logistics trajectory data based on a big data technology and a machine learning algorithm, and constructing a corresponding logistics trajectory prediction model; acquiring real-time logistics data and inputting the real-time logistics data into the logistics trajectory prediction model to obtain a corresponding trajectory prediction result; and establishing a corresponding logistics track optimization model according to a track prediction result in combination with logistics transportation requirements, and solving the logistics track optimization model to obtain an optimal logistics transportation path and a scheduling scheme, so that the logistics track can be accurately predicted and optimized, and the efficiency and reliability of logistics transportation are improved.

Description

Technical Field

[0001] The present invention relates to the field of logistics and transportation management technology, and in particular to a big data driven logistics trajectory intelligent prediction and path optimization system and method. Background Art

[0002] With the rapid development of information technology, big data technology is increasingly used in all walks of life, especially in the field of logistics. The development of big data technology has promoted the intelligence and efficiency of the logistics industry and accelerated the development of logistics trajectory prediction.

[0003] With the continuous development of advanced technologies such as cloud computing, the Internet of Things, and artificial intelligence, big data technology can better process massive amounts of data and achieve rapid analysis and decision-making. It has expanded from a single link to the entire supply chain. By integrating information from multiple logistics links, it can provide more comprehensive and accurate data support to help companies optimize operations and improve efficiency. Through big data technology, companies can transform large amounts of data into valuable information to guide decision-making and improve business processes. This digital transformation not only helps to enhance the competitiveness of companies, helps companies achieve more efficient and flexible resource allocation and transportation management, but also provides customers with a better service experience.

[0004] When processing massive trajectory data sets, the existing trajectory prediction methods have a sharp increase in computational complexity and low execution efficiency. Current data processing technologies are often unable to efficiently analyze logistics trajectory data, mine logistics transportation rules, trajectory information, etc., and provide key auxiliary decision-making assistance. In addition, existing logistics trajectory prediction optimization methods usually find it difficult to simultaneously take into account real-time performance and accuracy, and cannot meet the logistics industry's demand for rapid response.

[0005] Therefore, it is necessary to provide a big data-driven logistics trajectory intelligent prediction and path optimization system and method to solve the above technical problems. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a big data driven logistics trajectory intelligent prediction and path optimization system and method for solving the problems that the existing trajectory prediction method has a sharp increase in calculation amount and low execution efficiency when processing massive trajectory data sets, and it is unable to efficiently analyze logistics trajectory data, mine the logistics transportation laws, trajectory information, etc., and provide key auxiliary decision-making assistance. In addition, it is difficult to simultaneously take into account the real-time and accuracy of logistics trajectory prediction optimization, and cannot meet the logistics industry's demand for rapid response. The big data driven logistics trajectory intelligent prediction and path optimization system provided by the present invention comprises: A data collection module, used to collect historical logistics trajectory data, and the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; A data preprocessing module, used to perform preprocessing operations on the historical logistics trajectory data to generate corresponding target logistics trajectory data; A model building module, used to train and verify the target logistics trajectory data based on big data technology and machine learning algorithms, and build a corresponding logistics trajectory prediction model; A trajectory prediction module is used to obtain real-time logistics data and input it into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; The optimization decision module is used to establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and in combination with the logistics transportation demand, and solve the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan.

[0007] Preferably, the transport logistics trajectory data includes vehicle travel routes, vehicle travel time and transported goods types; the real-time traffic information includes road congestion conditions and traffic control information; and the weather condition data includes rainfall changes, snowfall changes and wind conditions.

[0008] Preferably, the data preprocessing module is used to perform preprocessing operations on the historical logistics trajectory data to generate corresponding target logistics trajectory data, specifically including: a cleaning unit, used to identify and delete duplicate, invalid and abnormal data records in the historical logistics trajectory data, and generate corresponding first logistics trajectory data; an integration unit, used for integrating the first logistics track data from different data sources to form unified second logistics track data; A standardization unit is used to perform standardization processing on the second logistics trajectory data, including standardization of date format and value range, to generate the target logistics trajectory data.

[0009] Preferably, the model building module is used to train and verify the target logistics trajectory data based on big data technology and machine learning algorithms to build a corresponding logistics trajectory prediction model, specifically including: Dividing the target logistics trajectory data into training logistics trajectory data and verification logistics trajectory data; Based on the training logistics trajectory data, a machine learning algorithm is used to construct the logistics trajectory prediction model; The target logistics trajectory features corresponding to the target logistics trajectory data are extracted, and the target logistics trajectory features are used as the input of the output layer neurons corresponding to the logistics trajectory prediction model. The calculation formula of the input of the output layer neurons is: In the formula, Indicates The input of the layer neurons is the input of the output layer neurons; Indicates The weight matrix of the neurons in the layer, that is, the weight matrix of the neurons in the output layer; Indicates The input of the neurons in the first layer is the activation value of the neurons in the previous layer of the output layer. Indicates The bias vector of the layer neurons, that is, the bias vector of the output layer neurons; The corresponding activation value of the output layer neuron is: In the formula, Indicates The activation value of the layer neurons, that is, the activation value of the output layer neurons; Represents the activation function corresponding to the logistics trajectory prediction model; Indicates The input of the neurons in the first layer is the input of the neurons in the output layer.

[0010] Preferably, actual training trajectory data in the training logistics trajectory data is obtained, corresponding predicted training trajectory data is determined according to the activation value of the output layer neuron, and based on the actual training trajectory data and the predicted training trajectory data, the loss function corresponding to the logistics trajectory prediction model is calculated: In the formula, Represents the loss function corresponding to the logistics trajectory prediction model; Indicates the number of data groups corresponding to the actual training trajectory data and the predicted training trajectory data; Indicates Actual training trajectory data; Indicates prediction training trajectory data; ∑ represents the summation symbol; The weight matrix and bias vector of the output layer neurons corresponding to the logistics trajectory prediction model are updated based on the gradient descent method. The updated weight matrix of the output layer neurons is: In the formula, Indicates the updated The weight matrix of the layer neurons, that is, the weight matrix of the output layer neurons after the update; Indicates The weight matrix of the neurons in the layer, that is, the weight matrix of the neurons in the output layer; Represents the learning rate corresponding to the logistics trajectory prediction model; Represents the loss function corresponding to the logistics trajectory prediction model; Represents the partial derivative of the loss function corresponding to the logistics trajectory prediction model with respect to the weight matrix of the output layer neurons; The updated bias vector of the output layer neuron is: In the formula, Indicates the updated The bias vector of the layer neurons, that is, the bias vector of the updated output layer neurons; Indicates The bias vector of the layer neurons, that is, the bias vector of the output layer neurons; Represents the learning rate corresponding to the logistics trajectory prediction model; Represents the loss function corresponding to the logistics trajectory prediction model; Represents the partial derivative of the loss function corresponding to the logistics trajectory prediction model with respect to the bias vector of the output layer neuron.

[0011] Preferably, after obtaining the logistics trajectory prediction model, the prediction effect of the logistics trajectory prediction model is evaluated based on the verification logistics trajectory data, and the logistics trajectory prediction model is adjusted according to the prediction effect, specifically including: Obtain a preset trajectory similarity threshold, if the similarity between the actual verification trajectory data in the verification logistics trajectory data and the corresponding predicted verification trajectory data is greater than or equal to the preset trajectory similarity threshold, then the verification logistics trajectory data is used as a positive sample; otherwise, if the similarity between the actual verification trajectory data in the verification logistics trajectory data and the corresponding predicted verification trajectory data is less than the preset trajectory similarity threshold, then the verification logistics trajectory data is used as a negative sample; Determine a confusion matrix corresponding to the verification logistics trajectory data, wherein the elements of the confusion matrix include true positive examples, false positive examples, true negative examples, and false negative examples, wherein the true positive examples are correctly predicted as positive samples; the false positive examples are incorrectly predicted as positive samples; the true negative examples are correctly predicted as negative samples; and the false negative examples are incorrectly predicted as negative samples; Based on the confusion matrix, the prediction indicators corresponding to the logistics trajectory prediction model are calculated, wherein the prediction indicators include accuracy, precision, recall, F1 score and false positive rate; The calculation formula of the accuracy is: In the formula, Indicates the accuracy of the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a true negative example, that is, a negative sample that is correctly predicted; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; Represents a false negative example, that is, a negative sample that is incorrectly predicted; The calculation formula of the accuracy is: In the formula, Indicates the accuracy of the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; The calculation formula of the recall rate is: In the formula, Indicates the recall rate corresponding to the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a false negative example, that is, a negative sample that is incorrectly predicted; The calculation formula of the F1 score is: In the formula, Indicates the F1 score corresponding to the logistics trajectory prediction model; Indicates the accuracy of the logistics trajectory prediction model; Indicates the recall rate corresponding to the logistics trajectory prediction model; The calculation formula of the false positive rate is: In the formula, Indicates the false positive rate corresponding to the logistics trajectory prediction model; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; Represents a true negative example, that is, a negative sample that is correctly predicted; The prediction indicators are analyzed, and the precision and recall are weighed to evaluate the prediction effect of the logistics trajectory prediction model, and the logistics trajectory prediction model is adjusted according to the prediction effect.

[0012] Preferably, the optimization decision module is used to establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and in combination with logistics transportation demand, and solve the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan, specifically including: A determination unit, configured to determine the corresponding logistics trajectory optimization target and constraint conditions according to the trajectory prediction result and in combination with the logistics transportation demand, wherein the logistics trajectory optimization target is to minimize the logistics transportation cost; An establishing unit, used for establishing the logistics trajectory optimization model based on a linear programming method according to the logistics trajectory optimization objective and constraint conditions; The result unit is used to solve the logistics trajectory optimization model based on the simplex method by iterating, searching and updating the decision variables corresponding to the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan.

[0013] A big data-driven logistics trajectory intelligent prediction and path optimization method, the method comprising: Collecting historical logistics trajectory data, wherein the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; Performing a preprocessing operation on the historical logistics trajectory data to generate corresponding target logistics trajectory data; Based on big data technology and machine learning algorithms, the target logistics trajectory data is trained and verified to build a corresponding logistics trajectory prediction model; Acquire real-time logistics data and input it into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; According to the trajectory prediction results and in combination with logistics transportation demand, a corresponding logistics trajectory optimization model is established, and the logistics trajectory optimization model is solved to obtain the optimal logistics transportation path and scheduling plan.

[0014] Compared with related technologies, the big data-driven logistics trajectory intelligent prediction and path optimization system and method provided by the present invention have the following beneficial effects: The present invention can collect historical logistics trajectory data through a data acquisition module, a data preprocessing module, a model building module, a trajectory prediction module and an optimization decision module, and the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; preprocessing operations are performed on the historical logistics trajectory data to generate corresponding target logistics trajectory data; based on big data technology and machine learning algorithms, the target logistics trajectory data is trained and verified to build a corresponding logistics trajectory prediction model; real-time logistics data is obtained and input into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; according to the trajectory prediction results, combined with logistics transportation needs, a corresponding logistics trajectory optimization model is established, and the logistics trajectory optimization model is solved to obtain the optimal logistics transportation path and scheduling plan. The present invention can accurately predict the logistics trajectory through big data technology and machine learning algorithms, improve the prediction accuracy and reliability; and can generate the optimal transportation path and scheduling plan based on the trajectory prediction results and in combination with multiple factors, reduce transportation costs, and improve transportation efficiency; at the same time, the system of the present invention can obtain and analyze traffic information, weather conditions and other data in real time, adjust the transportation plan in time, and enhance the ability to respond to emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a system block diagram of the big data driven logistics trajectory intelligent prediction and path optimization system of the present invention; Figure 2 It is a structural block diagram of the data preprocessing module of the big data driven logistics trajectory intelligent prediction and path optimization system of the present invention; Figure 3 It is a structural block diagram of the optimization decision module of the big data driven logistics trajectory intelligent prediction and path optimization system of the present invention; Figure 4 This is a flow chart of the big data driven logistics trajectory intelligent prediction and path optimization method of the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0017] Embodiment 1 like Figure 1 As shown, a big data driven logistics trajectory intelligent prediction and path optimization system, the system includes: A data collection module, used to collect historical logistics trajectory data, and the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; A data preprocessing module, used to perform preprocessing operations on the historical logistics trajectory data to generate corresponding target logistics trajectory data; A model building module, used to train and verify the target logistics trajectory data based on big data technology and machine learning algorithms, and build a corresponding logistics trajectory prediction model; A trajectory prediction module is used to obtain real-time logistics data and input it into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; The optimization decision module is used to establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and in combination with the logistics transportation demand, and solve the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan.

[0018] Among them, the data collection module can comprehensively and efficiently collect historical logistics trajectory data. These historical logistics trajectory data not only include the complete transportation logistics trajectory data from the starting point to the destination, that is, the specific path record of each logistics activity in time and space, but also integrate real-time traffic information and weather conditions data. Real-time traffic information includes road congestion, traffic accident reports and traffic control measures, which directly affect the efficiency and safety of logistics transportation, while weather conditions data include meteorological factors such as temperature, humidity, rainfall and wind speed. These factors may have an important impact on the logistics transportation process. For example, heavy rain may cause waterlogging on the road and reduced visibility, which in turn affects the driving speed and safety of logistics vehicles.

[0019] The data preprocessing module can perform preprocessing operations on the collected historical logistics trajectory data to ensure the accuracy and reliability of subsequent model construction. Preprocessing operations include data cleaning, data format integration, and data standardization, so as to generate more standardized and high-quality target logistics trajectory data, providing a solid foundation for subsequent analysis.

[0020] The model building module can perform in-depth mining and training on the pre-processed target logistics trajectory data based on big data technology and machine learning algorithms. Through the training process, the complex relationship between different logistics trajectories and various influencing factors can be learned, and the corresponding logistics trajectory prediction model can be built. The model can predict the future logistics trajectory trend based on the input data characteristics.

[0021] The trajectory prediction module can obtain real-time logistics data, input these data into the already constructed logistics trajectory prediction model, and output the corresponding trajectory prediction results, providing a scientific basis for the real-time scheduling of logistics transportation.

[0022] Finally, the optimization decision module can establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and comprehensively consider the current logistics transportation needs, such as cargo type, urgency, and customer requirements. By solving this model, the optimal logistics transportation path and scheduling plan can be obtained to achieve the goal of minimizing logistics transportation costs, thereby greatly improving the intelligence level of logistics management and overall operational efficiency.

[0023] During the specific implementation process, the transportation logistics trajectory data includes vehicle driving routes, vehicle driving time and transported goods types; the real-time traffic information includes road congestion and traffic control information; the weather condition data includes rainfall changes, snowfall changes and wind conditions.

[0024] It can be understood that transportation logistics trajectory data includes vehicle driving routes, vehicle driving times and types of transported goods, which provide basic information for logistics management and analysis; real-time traffic information includes road congestion and traffic control information, which ensures the optimization of transportation routes and efficient scheduling; weather condition data includes rainfall changes, snowfall changes and wind conditions, which have an important impact on logistics transportation safety and efficiency.

[0025] like Figure 2 As shown, the data preprocessing module 2 is used to perform preprocessing operations on the historical logistics trajectory data to generate corresponding target logistics trajectory data, specifically including: A cleaning unit 21, used to identify and delete duplicate, invalid and abnormal data records in the historical logistics trajectory data, and generate corresponding first logistics trajectory data; An integration unit 22, used to integrate the first logistics track data from different data sources to form unified second logistics track data; The standardization unit 23 is used to perform standardization processing on the second logistics trajectory data, including standardization of date format and value range, to generate the target logistics trajectory data.

[0026] It should be noted that for the data preprocessing module, the cleaning unit can identify and delete duplicate, invalid and abnormal data records in the historical logistics trajectory data. Specifically, duplicate data may be generated due to system errors or repeated data entry; invalid data may include null values, missing values ​​or obviously illogical records; abnormal data refers to values ​​that significantly deviate from the normal range, and the existence of these data may seriously affect the accuracy of subsequent analysis. Through the processing of the cleaning unit, the corresponding first logistics trajectory data can be obtained.

[0027] Next, the integration unit can integrate the first logistics track data from different data sources. In actual applications, logistics data may come from multiple different systems or platforms, such as warehouse management systems, transportation management systems, etc., and these data may differ in format, coding standards, etc. The integration unit can integrate these heterogeneous data into unified second logistics track data through technical means such as data mapping and format conversion.

[0028] Finally, the standardization unit can perform standardization on the second logistics trajectory data, including the unification of the date format and the standardization of the numerical range. Specifically, the unification of the date format helps to eliminate the analysis barriers caused by different time representations; while the standardization of the numerical range can convert data of different dimensions and scales into a comparable form, thereby generating the corresponding target logistics trajectory data.

[0029] The model building module is used to train and verify the target logistics trajectory data based on big data technology and machine learning algorithms to build a corresponding logistics trajectory prediction model, specifically including: Dividing the target logistics trajectory data into training logistics trajectory data and verification logistics trajectory data; Based on the training logistics trajectory data, a machine learning algorithm is used to construct the logistics trajectory prediction model; The target logistics trajectory features corresponding to the target logistics trajectory data are extracted, and the target logistics trajectory features are used as the input of the output layer neurons corresponding to the logistics trajectory prediction model. The calculation formula of the input of the output layer neurons is: In the formula, Indicates The input of the layer neurons is the input of the output layer neurons; Indicates The weight matrix of the neurons in the layer, that is, the weight matrix of the neurons in the output layer; Indicates The input of the neurons in the first layer is the activation value of the neurons in the previous layer of the output layer. Indicates The bias vector of the layer neurons, that is, the bias vector of the output layer neurons; The corresponding activation value of the output layer neuron is: In the formula, Indicates The activation value of the layer neurons, that is, the activation value of the output layer neurons; Represents the activation function corresponding to the logistics trajectory prediction model; Indicates The input of the neurons in the first layer is the input of the neurons in the output layer.

[0030] Among them, the model building module can first divide the collected target logistics trajectory data into training logistics trajectory data and verification logistics trajectory data, so as to ensure the data independence and objectivity of the verification link in the model training process, which helps to improve the generalization ability and prediction accuracy of the model. Based on the divided training logistics trajectory data, the model building module can use machine learning algorithms to automatically learn the potential laws and features in the training logistics trajectory data, and gradually build a logistics trajectory prediction model through iterative optimization.

[0031] It should be noted that the activation function determines the nonlinear processing capability of the logistics trajectory prediction model for the input data. It can flexibly adjust the output value according to the characteristics of the input data, so that the logistics trajectory prediction model can better fit the complex data distribution.

[0032] The actual training trajectory data in the training logistics trajectory data is obtained, the corresponding predicted training trajectory data is determined according to the activation value of the output layer neuron, and the loss function corresponding to the logistics trajectory prediction model is calculated based on the actual training trajectory data and the predicted training trajectory data: In the formula, Represents the loss function corresponding to the logistics trajectory prediction model; Indicates the number of data groups corresponding to the actual training trajectory data and the predicted training trajectory data; Indicates Actual training trajectory data; Indicates prediction training trajectory data; ∑ represents the summation symbol; The weight matrix and bias vector of the output layer neurons corresponding to the logistics trajectory prediction model are updated based on the gradient descent method. The updated weight matrix of the output layer neurons is: In the formula, Indicates the updated The weight matrix of the layer neurons, that is, the weight matrix of the output layer neurons after the update; Indicates The weight matrix of the neurons in the layer, that is, the weight matrix of the neurons in the output layer; Represents the learning rate corresponding to the logistics trajectory prediction model; Represents the loss function corresponding to the logistics trajectory prediction model; Represents the partial derivative of the loss function corresponding to the logistics trajectory prediction model with respect to the weight matrix of the output layer neurons; The updated bias vector of the output layer neuron is: In the formula, Indicates the updated The bias vector of the layer neurons, that is, the bias vector of the updated output layer neurons; Indicates The bias vector of the layer neurons, that is, the bias vector of the output layer neurons; Represents the learning rate corresponding to the logistics trajectory prediction model; Represents the loss function corresponding to the logistics trajectory prediction model; Represents the partial derivative of the loss function corresponding to the logistics trajectory prediction model with respect to the bias vector of the output layer neuron.

[0033] It should be noted that the logistics trajectory prediction model can continuously adjust its internal parameters to more accurately predict the logistics trajectory, and this process is an iterative optimization cycle. When the loss function reaches the preset threshold or the number of iterations reaches the preset upper limit, the training of the logistics trajectory prediction model is completed.

[0034] After obtaining the logistics trajectory prediction model, the prediction effect of the logistics trajectory prediction model is evaluated based on the verification logistics trajectory data, and the logistics trajectory prediction model is adjusted according to the prediction effect, specifically including: Obtain a preset trajectory similarity threshold, if the similarity between the actual verification trajectory data in the verification logistics trajectory data and the corresponding predicted verification trajectory data is greater than or equal to the preset trajectory similarity threshold, then the verification logistics trajectory data is used as a positive sample; otherwise, if the similarity between the actual verification trajectory data in the verification logistics trajectory data and the corresponding predicted verification trajectory data is less than the preset trajectory similarity threshold, then the verification logistics trajectory data is used as a negative sample; Determine a confusion matrix corresponding to the verification logistics trajectory data, wherein the elements of the confusion matrix include true positive examples, false positive examples, true negative examples, and false negative examples, wherein the true positive examples are correctly predicted as positive samples; the false positive examples are incorrectly predicted as positive samples; the true negative examples are correctly predicted as negative samples; and the false negative examples are incorrectly predicted as negative samples; Based on the confusion matrix, the prediction indicators corresponding to the logistics trajectory prediction model are calculated, wherein the prediction indicators include accuracy, precision, recall, F1 score and false positive rate; The calculation formula of the accuracy is: In the formula, Indicates the accuracy of the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a true negative example, that is, a negative sample that is correctly predicted; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; Represents a false negative example, that is, a negative sample that is incorrectly predicted; The calculation formula of the accuracy is: In the formula, Indicates the accuracy of the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; The calculation formula of the recall rate is: In the formula, Indicates the recall rate corresponding to the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a false negative example, that is, a negative sample that is incorrectly predicted; The calculation formula of the F1 score is: In the formula, Indicates the F1 score corresponding to the logistics trajectory prediction model; Indicates the accuracy of the logistics trajectory prediction model; Indicates the recall rate corresponding to the logistics trajectory prediction model; The calculation formula of the false positive rate is: In the formula, Indicates the false positive rate corresponding to the logistics trajectory prediction model; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; Represents a true negative example, that is, a negative sample that is correctly predicted; The prediction indicators are analyzed, and the precision and recall are weighed to evaluate the prediction effect of the logistics trajectory prediction model, and the logistics trajectory prediction model is adjusted according to the prediction effect.

[0035] It is understandable that in order to quantify the degree of closeness between the predicted trajectory and the actual trajectory, a trajectory similarity threshold can be pre-set first. The selection of this threshold needs to be based on a comprehensive consideration of business needs and data characteristics to distinguish between the acceptable range and the unacceptable range of the prediction results. Subsequently, the actual verification trajectory data in the verification logistics trajectory data can be compared with the corresponding predicted verification trajectory data to calculate the similarity between them. If the similarity is greater than or equal to the preset trajectory similarity threshold, the verification logistics trajectory data is used as a positive sample; conversely, if the similarity is less than the preset trajectory similarity threshold, the verification logistics trajectory data is used as a negative sample.

[0036] In order to deeply analyze the prediction performance of the logistics trajectory prediction model, the corresponding confusion matrix can be constructed. Based on the confusion matrix, key prediction indicators can be further calculated to comprehensively evaluate the performance of the logistics trajectory prediction model. These indicators include accuracy, precision, recall, F1 score and false positive rate. Specifically, the accuracy reflects the overall prediction ability of the logistics trajectory prediction model; the precision measures the recognition accuracy of the logistics trajectory prediction model; the recall reflects the coverage ability of the logistics trajectory prediction model; the F1 score is the harmonic mean of the precision and recall, which is used to weigh the two and provide a comprehensive performance indicator; the false positive rate reveals the possible false positives of the logistics trajectory prediction model.

[0037] It should be noted that a high accuracy rate usually means that the overall performance of the logistics trajectory prediction model is good, but it may mask the imbalance problem in some categories; a high precision rate indicates that the logistics trajectory prediction model has a strong ability to identify positive samples, but it may sacrifice the recall rate; a high recall rate means that the logistics trajectory prediction model can capture more positive samples, but it may be accompanied by a higher false positive rate. Therefore, when evaluating the prediction effect of the logistics trajectory prediction model, it is necessary to comprehensively consider these indicators, especially to find the best balance between precision and recall rate to meet the needs of practical applications.

[0038] In practical applications, the logistics trajectory prediction model can be adjusted accordingly according to the prediction results, including adjusting model parameters, optimizing feature selection, introducing new algorithms, etc., so as to improve the prediction performance of the logistics trajectory prediction model and make it more in line with the requirements of actual application scenarios.

[0039] like Figure 3 As shown, the optimization decision module 3 is used to establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and combined with the logistics transportation demand, and solve the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan, specifically including: A determination unit 31 is used to determine the corresponding logistics trajectory optimization target and constraint conditions according to the trajectory prediction result and the logistics transportation demand, and the logistics trajectory optimization target is to minimize the logistics transportation cost; An establishing unit 32, configured to establish the logistics trajectory optimization model based on a linear programming method according to the logistics trajectory optimization objective and constraint conditions; The result unit 33 is used to solve the logistics trajectory optimization model based on the simplex method by iterating, searching and updating the decision variables corresponding to the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan.

[0040] It can be understood that the optimization decision-making module can construct and solve the logistics trajectory optimization model based on the trajectory prediction results and specific logistics transportation needs, and optimize the logistics transportation path and scheduling plan through intelligent means, thereby significantly improving logistics transportation efficiency and reducing operating costs.

[0041] Specifically, the determination unit can determine the logistics trajectory optimization target, that is, minimizing the logistics transportation cost, based on the trajectory prediction results and specific logistics transportation needs. These logistics transportation needs include the type, quantity, urgency and customer location of the goods, etc. At the same time, corresponding constraints need to be set, such as time window restrictions, vehicle load restrictions and driver working hours regulations.

[0042] The establishment unit can use the linear programming method to construct a logistics trajectory optimization model according to the logistics trajectory optimization objectives and constraints output by the determination unit. It should be noted that linear programming is a mathematical optimization technique that is particularly suitable for dealing with problems with linear relationships such as resource allocation and path selection. The logistics trajectory optimization model can define decision variables, objective functions, and related constraints in detail, where decision variables can be vehicle allocation, path selection, etc.

[0043] The result unit can use the simplex method to continuously search and update the decision variables in the logistics trajectory optimization model through an iterative process until the optimal solution that meets all constraints and minimizes the logistics transportation cost is found. In this way, the optimal solution can be efficiently selected from many possible paths and scheduling solutions, providing logistics companies with accurate and efficient transportation guidance.

[0044] Embodiment 2 like Figure 4 As shown, a big data driven logistics trajectory intelligent prediction and path optimization method includes: S1, collecting historical logistics trajectory data, and the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; S2, performing a preprocessing operation on the historical logistics trajectory data to generate corresponding target logistics trajectory data; S3, based on big data technology and machine learning algorithms, training and verifying the target logistics trajectory data, and building a corresponding logistics trajectory prediction model; S4, acquiring real-time logistics data and inputting it into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; S5, according to the trajectory prediction result and in combination with the logistics transportation demand, a corresponding logistics trajectory optimization model is established, and the logistics trajectory optimization model is solved to obtain the optimal logistics transportation path and scheduling plan.

[0045] Through the introduction of the above embodiments, the present invention uses a big data driven logistics trajectory intelligent prediction and path optimization system and method, through a data acquisition module, a data preprocessing module, a model building module, a trajectory prediction module and an optimization decision module, to collect historical logistics trajectory data, and the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; preprocessing operations are performed on the historical logistics trajectory data to generate corresponding target logistics trajectory data; based on big data technology and machine learning algorithms, the target logistics trajectory data is trained and verified to build a corresponding logistics trajectory prediction model; real-time logistics data is obtained and input into the logistics trajectory prediction model to obtain the corresponding trajectory prediction results; according to the trajectory prediction results, combined with logistics transportation needs, a corresponding logistics trajectory optimization model is established, and the logistics trajectory optimization model is solved to obtain the optimal logistics transportation path and scheduling plan. The present invention can accurately predict the logistics trajectory through big data technology and machine learning algorithms, improve the prediction accuracy and reliability; and can generate the optimal transportation path and scheduling plan based on the trajectory prediction results and in combination with multiple factors, reduce transportation costs, and improve transportation efficiency; at the same time, the system of the present invention can obtain and analyze traffic information, weather conditions and other data in real time, adjust the transportation plan in time, and enhance the ability to respond to emergencies.

[0046] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0047] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0048] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. Big data driven logistics trajectory intelligent prediction and path optimization system, characterized by: The system comprises: A data collection module, used to collect historical logistics trajectory data, and the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; A data preprocessing module, used to perform preprocessing operations on the historical logistics trajectory data to generate corresponding target logistics trajectory data; A model building module, used to train and verify the target logistics trajectory data based on big data technology and machine learning algorithms, and build a corresponding logistics trajectory prediction model; A trajectory prediction module is used to obtain real-time logistics data and input it into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; The optimization decision module is used to establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and in combination with the logistics transportation demand, and solve the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan.

2. The big data driven logistics trajectory intelligent prediction and path optimization system according to claim 1 is characterized in that: The transport logistics trajectory data includes vehicle travel routes, vehicle travel time and transported goods types; the real-time traffic information includes road congestion and traffic control information; the weather condition data includes rainfall changes, snowfall changes and wind conditions.

3. The big data driven logistics trajectory intelligent prediction and path optimization system according to claim 1 is characterized in that: The data preprocessing module is used to perform preprocessing operations on the historical logistics trajectory data to generate corresponding target logistics trajectory data, specifically including: a cleaning unit, used to identify and delete duplicate, invalid and abnormal data records in the historical logistics trajectory data, and generate corresponding first logistics trajectory data; an integration unit, used for integrating the first logistics track data from different data sources to form unified second logistics track data; A standardization unit is used to perform standardization processing on the second logistics trajectory data, including standardization of date format and value range, to generate the target logistics trajectory data.

4. The big data driven logistics trajectory intelligent prediction and path optimization system according to claim 1 is characterized in that: The model building module is used to train and verify the target logistics trajectory data based on big data technology and machine learning algorithms to build a corresponding logistics trajectory prediction model, specifically including: Dividing the target logistics trajectory data into training logistics trajectory data and verification logistics trajectory data; Based on the training logistics trajectory data, a machine learning algorithm is used to construct the logistics trajectory prediction model; The target logistics trajectory features corresponding to the target logistics trajectory data are extracted, and the target logistics trajectory features are used as the input of the output layer neurons corresponding to the logistics trajectory prediction model. The calculation formula of the input of the output layer neurons is: In the formula, Indicates The input of the layer neurons is the input of the output layer neurons; Indicates The weight matrix of the neurons in the layer, that is, the weight matrix of the neurons in the output layer; Indicates The input of the neurons in the first layer is the activation value of the neurons in the previous layer of the output layer. Indicates The bias vector of the layer neurons, that is, the bias vector of the output layer neurons; The corresponding activation value of the output layer neuron is: In the formula, Indicates The activation value of the layer neurons, that is, the activation value of the output layer neurons; Represents the activation function corresponding to the logistics trajectory prediction model; Indicates The input of the neurons in the first layer is the input of the neurons in the output layer.

5. The big data driven logistics trajectory intelligent prediction and path optimization system according to claim 4 is characterized in that: The actual training trajectory data in the training logistics trajectory data is obtained, the corresponding predicted training trajectory data is determined according to the activation value of the output layer neuron, and the loss function corresponding to the logistics trajectory prediction model is calculated based on the actual training trajectory data and the predicted training trajectory data: In the formula, Represents the loss function corresponding to the logistics trajectory prediction model; Indicates the number of data groups corresponding to the actual training trajectory data and the predicted training trajectory data; Indicates Actual training trajectory data; Indicates prediction training trajectory data; ∑ represents the summation symbol; The weight matrix and bias vector of the output layer neurons corresponding to the logistics trajectory prediction model are updated based on the gradient descent method. The updated weight matrix of the output layer neurons is: In the formula, Indicates the updated The weight matrix of the layer neurons, that is, the weight matrix of the output layer neurons after the update; Indicates The weight matrix of the neurons in the layer, that is, the weight matrix of the neurons in the output layer; Represents the learning rate corresponding to the logistics trajectory prediction model; Represents the loss function corresponding to the logistics trajectory prediction model; Represents the partial derivative of the loss function corresponding to the logistics trajectory prediction model with respect to the weight matrix of the output layer neurons; The updated bias vector of the output layer neuron is: In the formula, Indicates the updated The bias vector of the layer neurons, that is, the bias vector of the updated output layer neurons; Indicates The bias vector of the layer neurons, that is, the bias vector of the output layer neurons; Represents the learning rate corresponding to the logistics trajectory prediction model; Represents the loss function corresponding to the logistics trajectory prediction model; Represents the partial derivative of the loss function corresponding to the logistics trajectory prediction model with respect to the bias vector of the output layer neuron.

6. The big data driven logistics trajectory intelligent prediction and path optimization system according to claim 5 is characterized in that: After obtaining the logistics trajectory prediction model, the prediction effect of the logistics trajectory prediction model is evaluated based on the verification logistics trajectory data, and the logistics trajectory prediction model is adjusted according to the prediction effect, specifically including: Obtain a preset trajectory similarity threshold, if the similarity between the actual verification trajectory data in the verification logistics trajectory data and the corresponding predicted verification trajectory data is greater than or equal to the preset trajectory similarity threshold, then the verification logistics trajectory data is used as a positive sample; otherwise, if the similarity between the actual verification trajectory data in the verification logistics trajectory data and the corresponding predicted verification trajectory data is less than the preset trajectory similarity threshold, then the verification logistics trajectory data is used as a negative sample; Determine a confusion matrix corresponding to the verification logistics trajectory data, wherein the elements of the confusion matrix include true positive examples, false positive examples, true negative examples, and false negative examples, wherein the true positive examples are correctly predicted as positive samples; the false positive examples are incorrectly predicted as positive samples; the true negative examples are correctly predicted as negative samples; and the false negative examples are incorrectly predicted as negative samples; Based on the confusion matrix, the prediction indicators corresponding to the logistics trajectory prediction model are calculated, wherein the prediction indicators include accuracy, precision, recall, F1 score and false positive rate; The calculation formula of the accuracy is: In the formula, Indicates the accuracy of the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a true negative example, that is, a negative sample that is correctly predicted; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; Represents a false negative example, that is, a negative sample that is incorrectly predicted; The calculation formula of the accuracy is: In the formula, Indicates the accuracy of the logistics trajectory prediction model; Represents a true positive example, that is, a correctly predicted positive sample; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; The calculation formula of the recall rate is: In the formula, Indicates the recall rate corresponding to the logistics trajectory prediction model; Represents a true positive example, that is, a positive sample that is correctly predicted; Represents a false negative example, that is, a negative sample that is incorrectly predicted; The calculation formula of the F1 score is: In the formula, Indicates the F1 score corresponding to the logistics trajectory prediction model; Indicates the accuracy of the logistics trajectory prediction model; Indicates the recall rate corresponding to the logistics trajectory prediction model; The calculation formula of the false positive rate is: In the formula, Indicates the false positive rate corresponding to the logistics trajectory prediction model; Represents a false positive example, that is, a sample that is incorrectly predicted to be positive; Represents a true negative example, that is, a negative sample that is correctly predicted; The prediction indicators are analyzed, and the precision and recall are weighed to evaluate the prediction effect of the logistics trajectory prediction model, and the logistics trajectory prediction model is adjusted according to the prediction effect.

7. The big data driven logistics trajectory intelligent prediction and path optimization system according to claim 1 is characterized in that: The optimization decision module is used to establish a corresponding logistics trajectory optimization model based on the trajectory prediction results and in combination with the logistics transportation demand, and solve the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan, specifically including: A determination unit, configured to determine the corresponding logistics trajectory optimization target and constraint conditions according to the trajectory prediction result and in combination with the logistics transportation demand, wherein the logistics trajectory optimization target is to minimize the logistics transportation cost; An establishing unit, used for establishing the logistics trajectory optimization model based on a linear programming method according to the logistics trajectory optimization objective and constraint conditions; The result unit is used to solve the logistics trajectory optimization model based on the simplex method by iterating, searching and updating the decision variables corresponding to the logistics trajectory optimization model to obtain the optimal logistics transportation path and scheduling plan.

8. A method for intelligent prediction and path optimization of logistics trajectories driven by big data, applied to the system for intelligent prediction and path optimization of logistics trajectories driven by big data as claimed in any one of claims 1 to 7, the method comprising: Collecting historical logistics trajectory data, wherein the historical logistics trajectory data includes transportation logistics trajectory data, real-time traffic information and weather condition data; Performing a preprocessing operation on the historical logistics trajectory data to generate corresponding target logistics trajectory data; Based on big data technology and machine learning algorithms, the target logistics trajectory data is trained and verified to build a corresponding logistics trajectory prediction model; Acquire real-time logistics data and input it into the logistics trajectory prediction model to obtain corresponding trajectory prediction results; According to the trajectory prediction results and in combination with logistics transportation demand, a corresponding logistics trajectory optimization model is established, and the logistics trajectory optimization model is solved to obtain the optimal logistics transportation path and scheduling plan.

Images