Logistics information data compression method for intelligent logistics transportation

Abstract

The invention belongs to the technical field of logistics transportation, and discloses a logistics information data compression method for intelligent logistics transportation, and the method comprises the following specific steps: 1, historical data collection, processing and feature engineering collection of historical data: obtaining historical logistics information from a logistics system; by systematically collecting, processing and analyzing historical logistics data and real-time vehicle information, efficient integration and utilization of data are achieved, firstly, key information is extracted through feature engineering, a high-precision cargo classification model is constructed, a solid foundation is provided for subsequent intelligent scheduling, and then, by combining the vehicle information and a cargo classification result, a cargo classification result is obtained. An intelligent scheduling model is constructed and optimized, optimal matching of vehicles and cargos is ensured, meanwhile, constraint conditions of safety and compliance are integrated, the safety and efficiency of the transportation process are further improved, and finally, logistics information data are efficiently compressed by adopting a lossless and lossy compression algorithm.

Description

Technical Field

[0001] The present invention belongs to the technical field of logistics transportation, and specifically relates to a logistics information data compression method for intelligent logistics transportation. Background Art

[0002] Intelligent logistics transportation is based on traditional logistics and uses advanced technologies such as the Internet of Things, big data, and AI to achieve the informatization and intelligence of logistics transportation. It realizes real-time monitoring and management of the transportation process, optimizes route planning, improves transportation efficiency and accuracy, reduces costs, and enhances customer satisfaction. The intelligent logistics transportation system can also achieve the automation, visualization, controllability, and networking of logistics, thereby improving resource utilization and productivity levels, and is an important means for the logistics industry to "reduce costs and increase efficiency".

[0003] With the continuous increase in the volume of logistics transportation, logistics information data has become extremely large and complex. In order to make logistics information more concise, it is necessary to compress logistics information data. Existing logistics information data compression generally focuses on the compression rate of data, while there is a certain degree of neglect of the relationship between cargo classification and vehicle scheduling. There is a close relationship between cargo classification and vehicle scheduling, and the two affect each other, which may lead to problems in logistics transportation efficiency and costs. When the cargo classification is inaccurate, it may lead to a mixture of cargo types loaded on the vehicle, which does not meet the transportation requirements or standards. This may cause the driver to frequently adjust the cargo position or reload during the transportation process, thereby prolonging the transportation time and reducing the transportation efficiency. At the same time, inaccurate cargo classification may cause the cargo to be squeezed, collided, or damaged during transportation, increasing the risk of cargo loss. And different types of cargo may require different transportation conditions and equipment, and inaccurate classification may lead to mismatched transportation equipment, increasing the safety risk during transportation. Therefore, it needs to be improved. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a logistics information data compression method for intelligent logistics transportation, which has the advantage of compressing based on cargo categories.

[0005] To achieve the above object, the present invention provides the following technical solution: A logistics information data compression method for intelligent logistics transportation, and the specific steps are as follows:

[0006] Step 1: Historical data collection, processing, and feature engineering

[0007] Collect historical data: Obtain historical logistics information from the logistics system, including cargo type, weight, volume, destination, shelf life, and transportation requirements;

[0008] Data preprocessing: Remove duplicate records, invalid data, and obviously incorrect data, and fill in missing values. Finally, convert data with different dimensions into a unified scale for subsequent processing and analysis;

[0009] Feature engineering: Extract key features, perform feature selection and transformation to improve the accuracy and generalization ability of subsequent models;

[0010] Step 2: Build a classification model:

[0011] Based on feature engineering, build a classification model so that the classification model can classify goods according to one or more features;

[0012] Step 3: Collect vehicle information

[0013] Collect vehicle information: such as vehicle ID, driver information, vehicle space, vehicle type, real-time status;

[0014] Structured processing: Establish a unified data standard to perform structured processing on vehicle information for subsequent data analysis and application;

[0015] Step 4: Build an intelligent scheduling model

[0016] Feature selection and extraction: Extract key vehicle features from vehicle information;

[0017] Model construction: Based on the classification model in Step 2 and the selected and extracted vehicle features, build a preliminary intelligent scheduling model;

[0018] Model optimization: Set the objective function to clarify the goal of model optimization, and use the goal as the direction for optimizing the intelligent scheduling model for subsequent training;

[0019] Consideration of constraints: Add necessary constraints to the model, which help ensure the safety and compliance of the transportation process;

[0020] Model training and validation: Integrate the constraints and the objective function, use historical data for model training, and verify the performance of the model through cross-validation. According to the verification results, make necessary adjustments to the model;

[0021] Step 5: Vehicle scheduling and route planning

[0022] Use the intelligent scheduling model to match vehicles and goods, and then use Internet of Things technology to monitor the status of goods and vehicles in real time, including the location of goods, vehicle status, and traffic conditions; According to the real-time data, use the route planning algorithm to plan the optimal transportation route for the vehicle, and dynamically adjust the transportation route according to the real-time location and status of the vehicle and the changes in traffic conditions during transportation to ensure transportation efficiency;

[0023] Step Six: Efficient Compression and Synchronous Update of Logistics Information Data

[0024] Lossless Compression of Important Data: The key attributes of goods and the key information of vehicles are regarded as important data. A lossless compression algorithm is used to compress the important data to ensure the integrity and accuracy of the data while reducing the data transmission and storage costs;

[0025] Lossy Compression of Unimportant Data: For some non-critical information, a lossy compression algorithm is adopted to sacrifice a certain data accuracy in exchange for a higher compression ratio, thereby reducing the data transmission and storage costs;

[0026] Data Synchronization and Real-time Update: An efficient data synchronization mechanism is established to ensure that the data in the logistics information system is consistent with the actual situation. Bidirectional data exchange between vehicles and the logistics information system is realized by using real-time communication technology; when the vehicle status, goods location or route planning changes, these changes are used as additional data and are updated to the logistics information system in real time together with the compressed logistics information.

[0027] Preferably, among the missing data and obvious error data described in Step One, for the missing data, interpolation method, mean substitution, and regression prediction methods are used to fill in according to specific situations; for the obvious error data, such as unreasonable weight, volume, and date, manual review is used for supplementation. If the manual review fails to supplement the data, then the data is automatically corrected by interpolation method, mean substitution, and regression prediction methods.

[0028] Preferably, the key features extracted by the feature engineering described in Step One include goods type, weight, volume, shelf life, transportation requirements, destination, and delivery time. When performing feature engineering, if the features contain categorical data, appropriate encoding processing is required, and the categorical data is converted into numerical data by using one-hot encoding or label encoding.

[0029] Preferably, the classification formula adopted by the classification model described in Step Two is:

[0030]

[0031] Preferably, when matching goods and vehicles in Step Five, on the same transportation route, the intelligent scheduling model will merge the goods with similar transportation conditions to reduce the transportation risks and costs caused by the differences in goods characteristics, and accurately calculate the space required for different goods through algorithms to maximize the vehicle loading rate and avoid unnecessary waste of the vehicle internal space; at the same time, when allocating goods, the intelligent scheduling model will comprehensively consider the carrying capacity of each vehicle and the total weight of the goods currently loaded on it.

[0032] Preferably, when matching goods with vehicles in step five, for goods that are vulnerable to damage, of high value, or require special handling, the intelligent scheduling model will automatically allocate vehicles with corresponding transportation qualifications and equipment, and adopt advanced packaging technologies and transportation strategies to ensure the safety and integrity of these goods during transportation.

[0033] Preferably, the calculation formula for the maximum loading rate of the vehicle described in step four is:

[0034]

[0035] Preferably, the key attributes of the goods described in step six include goods type, weight, volume, shelf life, and transportation requirements, and the key information of the vehicle includes vehicle ID, driver information, and real-time status.

[0036] Preferably, some non-critical information described in step six includes detailed path trajectory records, unnecessary high-frequency location updates based on the planned path, partial environmental monitoring data, and temperature and humidity changes during non-critical periods.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] By systematically collecting, processing, and analyzing historical logistics data and real-time vehicle information, the efficient integration and utilization of data are achieved. First, key information is extracted through feature engineering to construct a high-precision goods classification model, providing a solid foundation for subsequent intelligent scheduling. Then, combining vehicle information with the goods classification results, an intelligent scheduling model is constructed and optimized to ensure the optimal matching of vehicles and goods, and at the same time, constraints on safety and compliance are incorporated to further improve the safety and efficiency of the transportation process. Finally, lossless and lossy compression algorithms are used to efficiently compress logistics information data, which not only reduces the costs of data transmission and storage but also ensures the integrity and accuracy of the data. And by establishing an efficient data synchronization mechanism, it is ensured that the data in the logistics information system is consistent with the actual situation, realizing real-time two-way data exchange between the vehicle and the logistics information system, providing strong support for the intelligence and efficiency of logistics transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0041] The embodiments of the present invention provide: a method for compressing logistics information data for intelligent logistics transportation, and the specific steps are as follows:

[0042] Step 1: Historical Data Collection, Processing, and Feature Engineering

[0043] Collect historical data: Obtain historical logistics information from the logistics system, including cargo type, weight, volume, destination, shelf life, and transportation requirements;

[0044] Data preprocessing: Remove duplicate records, invalid data, and obviously incorrect data, as well as fill in missing values. Finally, convert data with different dimensions into a unified scale for subsequent processing and analysis;

[0045] Feature engineering: Extract key features, perform feature selection and transformation to improve the accuracy and generalization ability of subsequent models;

[0046] Step 2: Build a Classification Model

[0047] Based on feature engineering, build a classification model so that the classification model can classify cargo according to one or more features;

[0048] Step 3: Collect Vehicle Information

[0049] Collect vehicle information: Such as vehicle ID, driver information, vehicle space, vehicle type, and real-time status;

[0050] Structured processing: Establish a unified data standard to perform structured processing on vehicle information for subsequent data analysis and application;

[0051] Step 4: Build an Intelligent Scheduling Model

[0052] Feature selection and extraction: Extract key vehicle features from vehicle information;

[0053] Model construction: Based on the classification model in Step 2 and the selected and extracted vehicle features, build a preliminary intelligent scheduling model;

[0054] Model optimization: Set the objective function, clarify the goal of model optimization, and use the goal as the direction for optimizing the intelligent scheduling model for subsequent training;

[0055] Consideration of constraint conditions: Add necessary constraint conditions to the model, which helps to ensure the safety and compliance of the transportation process;

[0056] Model training and validation: Integrate constraint conditions and the objective function, use historical data for model training, and verify the performance of the model through cross-validation. According to the verification results, make necessary adjustments to the model;

[0057] Step 5: Vehicle Scheduling and Route Planning

[0058] Use an intelligent scheduling model to match vehicles and goods, and then use Internet of Things technology to monitor the status of goods and vehicles in real time, including the location of goods, vehicle status, and traffic conditions; according to the real-time data, use a path planning algorithm to plan the optimal transportation path for the vehicle, and dynamically adjust the transportation path according to the real-time location and status of the vehicle and the changes in traffic conditions during transportation to ensure transportation efficiency;

[0059] Step Six: Efficient Compression and Synchronous Update of Logistics Information Data

[0060] Lossless Compression of Important Data: The key attributes of goods and the key information of vehicles are regarded as important data, and a lossless compression algorithm is used to compress the important data to ensure the integrity and accuracy of the data while reducing data transmission and storage costs;

[0061] Lossy Compression of Unimportant Data: For some non-critical information, a lossy compression algorithm is used to sacrifice a certain data accuracy in exchange for a higher compression ratio, thereby reducing data transmission and storage costs;

[0062] Data Synchronization and Real-Time Update: Establish an efficient data synchronization mechanism to ensure that the data in the logistics information system is consistent with the actual situation, and use real-time communication technology to achieve two-way data exchange between vehicles and the logistics information system; when the vehicle status, goods location, or path planning changes, these changes are used as additional data and are updated to the logistics information system in real time together with the compressed logistics information.

[0063] First, collect and process historical logistics information, including the type, weight, volume, etc. of goods, and perform data preprocessing and feature engineering to improve the accuracy of the model. Then, build a classification model based on feature engineering to classify the goods, and at the same time collect vehicle information and perform structured processing. Then, build an intelligent scheduling model based on the classification model and vehicle characteristics, and set objective functions and constraint conditions for optimization training and verification. In the vehicle scheduling and path planning stage, use the intelligent scheduling model to match vehicles and goods, monitor the status in real time through Internet of Things technology, use the path planning algorithm to plan the optimal path, and dynamically adjust at the same time. Finally, perform efficient compression and synchronous update on the logistics information data, use lossless compression for important data, use lossy compression for unimportant data to reduce costs, and at the same time establish a data synchronization mechanism to ensure real-time data update.

[0064] Among them, for the missing data and the obvious error data in Step One, for the missing data, interpolation method, mean substitution, and regression prediction methods are used to fill in according to the specific situation; for the obvious error data, such as unreasonable weight, volume, and date, it is supplemented through manual review. If the manual review fails to supplement the data, then the data is automatically corrected through interpolation method, mean substitution, and regression prediction methods.

[0065] By correcting and filling in the missing data and the data with obvious errors, the integrity of the data is ensured, enabling the subsequent data processing to proceed smoothly.

[0066] Among them, the key features extracted by the feature engineering in step one include the type of goods, weight, volume, shelf life, transportation requirements, destination, and delivery time. When performing feature engineering, if the features contain categorical data, appropriate encoding processing is required. The categorical data is converted into numerical data by using one-hot encoding or label encoding.

[0067] Through feature engineering, the original data is transformed and extracted to make it closer to the target variable, thereby improving the prediction accuracy of the model. By extracting the features closely related to key indicators such as logistics efficiency and cost through feature engineering, a more accurate intelligent scheduling model is constructed.

[0068] Among them, the classification formula adopted by the classification model in step two is:

[0069]

[0070] Among them:

[0071] Class i represents the probability that the goods are classified into the i-th category, and finally the category with the highest probability is selected as the classification result;

[0072] C is the total number of categories;

[0073] σ is the activation function, such as the softmax function, which is used to convert the result of the linear combination into a probability distribution;

[0074] w ij is the weight of the j-th feature for the i-th category, and these weights are obtained by training the classification model;

[0075] f j (x) is the transformation function of the j-th feature, which can be a linear function or a non-linear function. Here, x represents the feature vector of the goods, including the type of goods, weight, volume, destination, shelf life, and transportation requirements;

[0076] b i is the bias term of the i-th category.

[0077] Among them, when matching goods with vehicles in step five, on the same transportation route, the intelligent scheduling model will merge goods with similar transportation conditions to reduce transportation risks and costs caused by differences in goods characteristics. And through algorithms, it precisely calculates the space required for different goods to maximize the vehicle loading rate while avoiding unnecessary waste of the vehicle interior space. At the same time, when allocating goods, the intelligent scheduling model will comprehensively consider the carrying capacity of each vehicle and the total weight of the goods currently loaded on it.

[0078] Through intelligent algorithms, the optimal allocation of goods is carried out to achieve the balance of weight distribution, improve transportation safety and efficiency, avoid overloading, and ensure the smooth progress of the transportation process.

[0079] Among them, when matching goods with vehicles in step five, for vulnerable, high-value or goods requiring special handling, the intelligent scheduling model will automatically allocate vehicles with corresponding transportation qualifications and equipment, and through the adoption of advanced packaging technologies and transportation strategies, to ensure the safety and integrity of these goods during transportation.

[0080] Through the special attention of the intelligent scheduling model to vulnerable, high-value or goods requiring special handling, the safety of such goods during sorting and transportation is ensured, enabling the goods to be delivered safely, intactly and on time. The intelligent scheduling system improves the reliability and punctuality of transportation by precisely matching goods and vehicles, thus enhancing customer satisfaction and trust.

[0081] Among them, in step four, the calculation formula for the maximum vehicle loading rate is:

[0082]

[0083] Where:

[0084] O represents the optimal loading plan;

[0085] L represents the set of loading plans;

[0086] v represents the vehicle;

[0087] Wv represents the total weight of the goods allocated to vehicle v;

[0088] V v represents the total volume of the goods allocated to vehicle v;

[0089] and respectively represent the maximum carrying weight and maximum carrying volume of vehicle v;

[0090] α and β are weight coefficients used to balance the utilization rates of weight and volume.

[0091] Among them, the key attributes of the goods in step six include the type of goods, weight, volume, shelf life, and transportation requirements, and the key information of the vehicle includes the vehicle ID, driver information, and real-time status.

[0092] Through these key information, the logistics transportation efficiency and transportation costs can be controlled and losslessly compressed, so as to ensure the accuracy of this key information.

[0093] Among them, some non-critical information in step six includes detailed path trajectory records, on the basis of the planned path, unnecessary high-frequency position updates, some environmental monitoring data, and temperature and humidity changes during non-critical periods.

[0094] By performing lossy compression on the non-critical information, the data transmission and storage costs can be reduced, and at the same time, the compression efficiency of the data can be improved.

[0095] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0096] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A logistics information data compression method for intelligent logistics transportation, characterized in that: The specific steps are as follows: Step 1: Historical data collection, processing and feature engineering Collect historical data: Obtain historical logistics information from the logistics system, including cargo type, weight, volume, destination, shelf life, and transportation requirements; Data preprocessing: remove duplicate records, invalid data, and obviously erroneous data, fill in missing values, and finally convert data of different dimensions into a unified scale to facilitate subsequent processing and analysis; Feature engineering: extract key features, perform feature selection and transformation to improve the accuracy and generalization ability of subsequent models; Step 2: Build a classification model: Based on feature engineering, a classification model is constructed so that the classification model can classify goods according to one or more features; Step 3: Collect vehicle information Collect vehicle information: such as vehicle ID, driver information, vehicle space, vehicle type, and real-time status; Structured processing: Establish unified data standards and perform structured processing on vehicle information to facilitate subsequent data analysis and application; Step 4: Build an intelligent scheduling model Feature selection and extraction: Extract key vehicle features from vehicle information; Model construction: Based on the classification model in step 2 and the selected and extracted vehicle features, a preliminary intelligent scheduling model is constructed; Model optimization: Set the objective function, clarify the goal of model optimization, and use the goal as the direction of intelligent scheduling model optimization so that it can be optimized in subsequent training; Constraint consideration: Add necessary constraints to the model, which help ensure the safety and compliance of the transportation process; Model training and verification: Comprehensive constraints and objective functions, use historical data to train the model, and verify the performance of the model through cross-validation. According to the verification results, make necessary adjustments to the model; Step 5: Vehicle Scheduling and Path Planning Use intelligent scheduling models to match vehicles and goods, and then use IoT technology to monitor the status of goods and vehicles in real time, including the location of goods, vehicle status, and traffic conditions. Use path planning algorithms to plan the optimal transportation path for vehicles based on real-time data. During transportation, dynamically adjust the transportation path based on the real-time location and status of the vehicle, as well as changes in traffic conditions, to ensure transportation efficiency. Step 6: Efficient compression and synchronous update of logistics information data Lossless compression of important data: The key attributes of the cargo and the key information of the vehicle are regarded as important data. The lossless compression algorithm is used to compress the important data to ensure that the integrity and accuracy of the data are maintained while reducing the data transmission and storage costs; Lossy compression of non-critical data: For some non-critical information, a lossy compression algorithm is used to sacrifice a certain amount of data accuracy in exchange for a higher compression ratio, thereby reducing data transmission and storage costs; Data synchronization and real-time update: Establish an efficient data synchronization mechanism to ensure that the data in the logistics information system is consistent with the actual situation, and use real-time communication technology to achieve two-way data exchange between vehicles and logistics information systems; when the vehicle status, cargo location or route planning changes, these changes are used as additional data and updated in real time to the logistics information system together with the compressed logistics information.

2. According to claim 1, a logistics information data compression method for intelligent logistics transportation is characterized by: For the missing data and obviously erroneous data described in step 1, the missing data shall be filled by interpolation, mean substitution, and regression prediction methods according to the specific circumstances; for obviously erroneous data, such as unreasonable weight, volume, and date, they shall be supplemented through manual review. If manual review fails to supplement the data, the data shall be automatically corrected by interpolation, mean substitution, and regression prediction methods.

3. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: The key features extracted by the feature engineering described in step 1 include cargo type, weight, volume, shelf life, transportation requirements, destination, and delivery time. When performing feature engineering, if the features contain categorical data, appropriate encoding processing is required. The categorical data is converted into numerical data by using one-hot encoding or label encoding.

4. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: The classification formula used by the classification model in step 2 is:

5. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: As described in step 5, when matching goods with vehicles, on the same transportation route, the intelligent scheduling model will merge goods with similar transportation conditions to reduce transportation risks and costs caused by differences in goods characteristics, and accurately calculate the space required for different goods through algorithms to maximize the vehicle loading rate while avoiding unnecessary waste of vehicle interior space; at the same time, when allocating goods, the intelligent scheduling model will comprehensively consider the carrying capacity of each vehicle and the total weight of its current cargo.

6. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: When matching goods with vehicles as described in step 5, for goods that are fragile, high-value, or require special handling, the intelligent scheduling model will automatically assign vehicles with corresponding transportation qualifications and equipment, and use advanced packaging technology and transportation strategies to ensure the safety and integrity of these goods during transportation.

7. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: The calculation formula for the maximum vehicle loading rate in step 4 is:

8. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: The key attributes of the goods described in step six include the type of goods, weight, volume, shelf life, and transportation requirements; the key information of the vehicle includes vehicle ID, driver information, and real-time status.

9. The method for compressing logistics information data for intelligent logistics transportation according to claim 1, characterized in that: Some non-critical information described in step six includes detailed path trajectory records, non-essential high-frequency location updates based on the planned path, some environmental monitoring data, and temperature and humidity changes during non-critical periods.

Images