Method, device, equipment and medium for optimizing lost express delivery

Abstract

The present invention provides a method, device, equipment and medium for optimizing express delivery loss, including: obtaining parcel data of each logistics link, and preprocessing the obtained parcel data of each logistics link; constructing an express delivery business data model according to an entity relationship model based on the preprocessed parcel data of each logistics link; analyzing risk factors of parcel loss based on the express delivery business data model; based on the preprocessed parcel data of each logistics link, according to the risk factors of parcel loss, using a risk warning model to predict the risk of parcel loss.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics optimization, and in particular to a method, device, equipment and medium for optimizing lost express delivery. Background Art

[0002] In the modern express logistics industry, the safety and timeliness of packages are key factors affecting customer satisfaction and corporate competitiveness. With the rapid development of e-commerce, the volume of express delivery business has increased dramatically, and the risk of package loss and damage has also increased. This not only brings inconvenience to consumers, but may also cause express delivery companies to suffer economic losses and reputation damage. Therefore, how to effectively reduce the risk of package loss and improve logistics management efficiency has become an urgent problem to be solved in the industry. Summary of the invention

[0003] The main purpose of the present invention is to solve the technical problems in the prior art that the risk of package loss cannot be effectively reduced and the efficiency of logistics management cannot be improved.

[0004] A first aspect of the present invention provides a method for optimizing lost express delivery, comprising: Obtain the parcel data of each logistics link, and pre-process the obtained parcel data of each logistics link; Based on the pre-processed parcel data of each logistics link, the express business data model is constructed according to the entity relationship model; Analyze the risk factors of package loss based on the express delivery business data model; Based on the pre-processed parcel data of each logistics link and the risk factors of parcel loss, the risk of parcel loss is predicted using the risk warning model.

[0005] Optionally, in a first implementation of the first aspect of the present invention, obtaining the package data of each logistics link includes: Obtain package data including collection, sorting, transportation and delivery through API interface; The temperature sensor, humidity sensor and vibration sensor installed on the package collect the status information of the package in real time; Get the size, weight and destination information of the package by scanning the barcode on the courier; The obtained package data of each logistics link is cleaned, feature extracted and standardized to obtain the pre-processed package data of each logistics link.

[0006] Optionally, in a second implementation of the first aspect of the present invention, analyzing the risk factors of package loss based on the express delivery service data model includes: Based on the express delivery business data model, the risk factors of package loss are analyzed through methods including data visualization, association rule mining and cluster analysis.

[0007] Optionally, in a third implementation of the first aspect of the present invention, the risk of package loss is predicted using a risk warning model based on the pre-processed package data of each logistics link according to the risk factors of package loss, including: Build risk warning models based on machine learning algorithms; Construct training sets and test sets for training risk warning models; The risk warning model is trained using the training set to obtain a trained risk warning model; Use the test set to evaluate the trained risk warning model, and further optimize the current risk warning model based on the evaluation results to obtain an optimized risk warning model; Using the optimized risk warning model, based on the pre-processed package data of each logistics link, it is predicted whether the package is at risk of loss. When there is a risk of loss, an early warning is triggered.

[0008] Optionally, in a fourth implementation of the first aspect of the present invention, the construction of a training set and a test set for training a risk warning model includes: Obtain the parcel data of historical logistics links, and pre-process the parcel data of historical logistics links, including: data cleaning, feature extraction and data standardization; Divide the pre-processed historical logistics parcel data into training set and test set; When the training set and / or the test set is an unbalanced data set, the unbalanced data set is mixed and sampled to obtain a balanced data set; The mixed sampling of the imbalanced data set includes: using an oversampling method to increase the number of minority class samples; and using an undersampling method to reduce the number of majority class samples.

[0009] Optionally, in a fifth implementation of the first aspect of the present invention, the method further comprises: analyzing historical data using a loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimizing the transportation route based on the predicted high-risk links and factors for package loss that meet preset requirements; The method of analyzing historical data using the loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimizing the transportation route based on the predicted high-risk links and factors for package loss that meet preset requirements, includes: Build a missing risk prediction model based on machine learning algorithms; Construct training and test sets for training loss risk prediction models based on historical data; Using the training set to train the loss risk prediction model, to obtain a trained loss risk prediction model; The trained loss risk prediction model is evaluated using the test set, and the current loss risk prediction model is further optimized according to the evaluation results to obtain an optimized loss risk prediction model; Based on the pre-processed parcel data of each logistics link, the parcel loss rate of different links is analyzed using the optimized loss risk prediction model. When the parcel loss rate is higher than the preset value, the current link is considered to be a high-risk link; For high-risk links, transportation routes are optimized using transportation route optimization models.

[0010] Optionally, in a sixth implementation of the first aspect of the present invention, optimizing the transportation route by using a transportation route optimization model for high-risk links includes: Based on the location information, destination, transportation time and transportation cost of the package as optimization objectives, and including road conditions and traffic restrictions as constraints, a transportation route optimization model is constructed; Genetic algorithms are used to optimize and solve the constructed transportation route optimization model to obtain the optimal transportation route.

[0011] A second aspect of the present invention provides a lost express optimization device, comprising: The data acquisition module is used to acquire the package data of each logistics link and pre-process the acquired package data of each logistics link; The business data model building module is used to build the express business data model based on the pre-processed parcel data of each logistics link according to the entity relationship model; The risk factor analysis module is used to analyze the risk factors of package loss based on the express delivery business data model; The risk warning module is used to predict the risk of package loss based on the pre-processed package data of each logistics link and the risk factors of package loss using the risk warning model.

[0012] Optionally, in a first implementation of the second aspect of the present invention, the data acquisition module includes: Obtain package data including collection, sorting, transportation and delivery through API interface; The temperature sensor, humidity sensor and vibration sensor installed on the package collect the status information of the package in real time; Get the size, weight and destination information of the package by scanning the barcode on the courier; The obtained package data of each logistics link is cleaned, feature extracted and standardized to obtain the pre-processed package data of each logistics link.

[0013] Optionally, in a second implementation of the second aspect of the present invention, the risk factor analysis module includes: Based on the express delivery business data model, the risk factors of package loss are analyzed through methods including data visualization, association rule mining and cluster analysis.

[0014] Optionally, in a third implementation of the second aspect of the present invention, the risk warning module includes: Build risk warning models based on machine learning algorithms; Construct training sets and test sets for training risk warning models; The risk warning model is trained using the training set to obtain a trained risk warning model; Use the test set to evaluate the trained risk warning model, and further optimize the current risk warning model based on the evaluation results to obtain an optimized risk warning model; Using the optimized risk warning model, based on the pre-processed package data of each logistics link, it is predicted whether the package is at risk of loss. When there is a risk of loss, an early warning is triggered.

[0015] Optionally, in a fourth implementation of the second aspect of the present invention, the construction of a training set and a test set for training a risk warning model includes: Obtain the parcel data of historical logistics links, and pre-process the parcel data of historical logistics links, including: data cleaning, feature extraction and data standardization; Divide the pre-processed historical logistics parcel data into training set and test set; When the training set and / or the test set is an unbalanced data set, the unbalanced data set is mixed and sampled to obtain a balanced data set; The mixed sampling of the imbalanced data set includes: using an oversampling method to increase the number of minority class samples; and using an undersampling method to reduce the number of majority class samples.

[0016] Optionally, in a fifth implementation of the second aspect of the present invention, the device further comprises: a transportation route optimization module, configured to analyze historical data using a loss risk prediction model, predict high-risk links and factors for package loss that meet preset requirements, and optimize the transportation route based on the predicted high-risk links and factors for package loss that meet preset requirements; The method of analyzing historical data using the loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimizing the transportation route based on the predicted high-risk links and factors for package loss that meet preset requirements, includes: Build a missing risk prediction model based on machine learning algorithms; Construct training and test sets for training loss risk prediction models based on historical data; Using the training set to train the loss risk prediction model, to obtain a trained loss risk prediction model; The trained loss risk prediction model is evaluated using the test set, and the current loss risk prediction model is further optimized according to the evaluation results to obtain an optimized loss risk prediction model; Based on the pre-processed parcel data of each logistics link, the parcel loss rate of different links is analyzed using the optimized loss risk prediction model. When the parcel loss rate is higher than the preset value, the current link is considered to be a high-risk link; For high-risk links, transportation routes are optimized using transportation route optimization models.

[0017] Optionally, in a sixth implementation of the second aspect of the present invention, optimizing the transportation route by using a transportation route optimization model for high-risk links includes: Based on the location information, destination, transportation time and transportation cost of the package as optimization objectives, and including road conditions and traffic restrictions as constraints, a transportation route optimization model is constructed; Genetic algorithms are used to optimize and solve the constructed transportation route optimization model to obtain the optimal transportation route.

[0018] A third aspect of the present invention provides an electronic device, the electronic device comprising a memory and at least one processor, the memory storing instructions; The at least one processor calls the instructions in the memory so that the electronic device executes the various steps of the express loss optimization method as described above.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, the various steps of the express delivery loss optimization method described above are implemented.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention provides a method, device, equipment and medium for optimizing lost express delivery, a closed-loop system from data collection, analysis, prediction, optimization to feedback, and a continuously iterated and improved solution, thereby effectively reducing the occurrence of lost express delivery events and improving overall work efficiency and customer satisfaction; 2. The present invention can significantly improve the optimization level of transportation routes, reduce unnecessary transfers and delays, thereby reducing the risk of loss and improving overall logistics efficiency; 3. The present invention helps to realize the intelligence and automation of the express delivery industry and improve the overall operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 A first flow chart of the express delivery loss optimization method provided in an embodiment of the present invention.

[0022] Figure 2 A second flow chart of the express loss optimization method provided in an embodiment of the present invention.

[0023] Figure 3 A third flow chart of the express loss optimization method provided in an embodiment of the present invention.

[0024] Figure 4 A schematic diagram of the structure of a lost express optimization device provided in an embodiment of the present invention.

[0025] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The embodiment of the present invention provides a method, device, equipment and medium for optimizing express delivery loss, which obtains the parcel data of each logistics link and pre-processes the parcel data of each logistics link; constructs an express delivery business data model based on the pre-processed parcel data of each logistics link according to the entity relationship model; analyzes the risk factors of parcel loss based on the express delivery business data model; and predicts the risk of parcel loss using a risk warning model based on the pre-processed parcel data of each logistics link according to the risk factors of parcel loss. The present invention solves the technical problem that the risk of parcel loss cannot be effectively reduced and the efficiency of logistics management cannot be improved in the prior art.

[0027] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1The first embodiment of the express delivery loss optimization method in the embodiment of the present invention includes: 101. Obtaining parcel data of each logistics link, and preprocessing the obtained parcel data of each logistics link; In this embodiment, the package data including the collection, sorting, transportation and delivery stages are obtained through the API interface; The temperature sensor, humidity sensor and vibration sensor installed on the package collect the status information of the package in real time; Get the size, weight and destination information of the package by scanning the barcode on the courier; The obtained package data of each logistics link is cleaned, feature extracted and standardized to obtain the pre-processed package data of each logistics link.

[0029] 102. Based on the pre-processed parcel data of each logistics link, the express business data model is constructed according to the entity relationship model; In this embodiment, based on the pre-processed parcel data of each logistics link, the entity relationship model is used to associate the information of the collection, sorting, transportation and delivery links to form a complete express delivery business data model.

[0030] 103. Analyze the risk factors of package loss based on the express delivery business data model; In this embodiment, based on the express delivery business data model, data analysis tools and algorithms are used, such as data visualization, association rule mining, and cluster analysis methods, to analyze package data and discover risk factors for package loss.

[0031] 104. Based on the pre-processed parcel data of each logistics link and the risk factors of parcel loss, the risk of parcel loss is predicted using the risk warning model.

[0032] In this embodiment, a risk warning model is constructed based on machine learning algorithms, such as LightGBM, random forest, and support vector machine; a training set and a test set for training the risk warning model are constructed; specifically, the parcel data of the historical logistics link is obtained, and the parcel data of the historical logistics link is preprocessed, including data cleaning, feature extraction, and data standardization processing; the preprocessed parcel data of the historical logistics link is divided into a training set and a test set; when the training set and / or the test set is an unbalanced data set, the unbalanced data set is mixed sampled to obtain a balanced data set; wherein, the mixed sampling of the unbalanced data set includes: using an oversampling method, such as SMOTE (SyntheticMinority Over-sampling Technique), ADASYN (Adaptive Synthetic Sampling), etc., to increase the number of minority class samples; using an undersampling method, such as random undersampling, Tomek links undersampling, etc., to reduce the number of majority class samples; through mixed sampling, the number of different categories in the sample set can be made roughly equal, thereby improving the classification performance of the model.

[0033] Use the LGBMClassifier class in the LightGBM library to set the model parameters, such as learning rate, tree depth, number of leaf nodes, etc. The risk warning model is trained using the training set. By continuously adjusting the model parameters, the accuracy and generalization ability of the model are improved to obtain the trained risk warning model. Use the test set to evaluate the trained risk warning model, including calculating the model's accuracy, recall rate, F1 value and other indicators. Further optimize the current risk warning model based on the evaluation results to obtain an optimized risk warning model; Using the optimized risk warning model, based on the pre-processed package data of each logistics link, it is predicted whether the package is at risk of loss. When there is a risk of loss, an early warning is triggered.

[0034] This embodiment can effectively reduce the risk of package loss. After reducing the risk of package loss, customer experience will be significantly improved, thereby increasing customer satisfaction.

[0035] See also Figure 2 The second embodiment of the express delivery loss optimization method in the embodiment of the present invention includes: 201. Obtaining parcel data of each logistics link, and preprocessing the obtained parcel data of each logistics link, and constructing an express business data model according to the entity relationship model based on the preprocessed parcel data of each logistics link; In this embodiment, information on the collection, sorting, transportation, and delivery of items is obtained through an API interface; Temperature sensors, humidity sensors, vibration sensors and other sensors are installed on express parcels to collect parcel status information in real time; Use the mobile app to scan the barcode on the package and enter the package's dimensions, weight, destination and other information.

[0036] Use a relational or NoSQL database to store information such as the package's dimensions, weight, destination, estimated time of arrival, actual status, etc. Use data cleaning tools or write scripts to clean and preprocess data to remove duplicate data, erroneous data, and incomplete data; Use the entity-relationship model to link information from collection, sorting, transportation, and delivery to form a complete express delivery business data model.

[0037] 202. Analyze the risk factors of package loss based on the express delivery business data model; 203. Based on the pre-processed parcel data of each logistics link, according to the risk factors of parcel loss, the risk of parcel loss is predicted using the risk warning model; In this embodiment, machine learning algorithms, such as LightGBM, random forest, support vector machine, etc., are used to establish a risk warning model to predict the risk of package loss based on factors such as the package's location information, status information, and transportation route; more specifically, express delivery business data is collected and organized, including the package's location information, status information, and transportation route. Divide the data into training set and test set for model training and evaluation, preprocess the data, including data cleaning, feature extraction, data standardization, etc., use data cleaning tools or write scripts to remove noise and outliers in the data; feature extraction algorithms such as principal component analysis (PCA) and linear discriminant analysis (LDA) can be used to extract features from the data; data standardization methods such as Z-score standardization and Min-Max standardization can be used to standardize the data to the same scale; perform mixed sampling on the unbalanced data set to obtain a balanced sample set, use oversampling methods such as SMOTE (Synthetic Minority Over-sampling Technique) and ADASYN (Adaptive Synthetic Sampling) to increase the number of minority class samples; use undersampling methods such as random undersampling and Tomek links undersampling to reduce the number of majority class samples. Through mixed sampling, the number of different categories in the sample set can be roughly equal, improving the classification performance of the model; use the LGBMClassifier class in the LightGBM library to set the parameters of the model, such as the learning rate, tree depth, number of leaf nodes, etc. Use the training set to train the model, and continuously adjust the model parameters to improve the model's accuracy and generalization ability; evaluate and optimize the trained model, use the test set to evaluate the model, and calculate the model's accuracy, recall rate, F1 value and other indicators. According to the evaluation results, adjust the model's parameters or select other algorithms to further optimize the model's performance.

[0038] Deploy the trained model to the production environment, monitor the location and status of the package in real time, and use real-time data processing frameworks such as Storm and Flink to input the location and status information of the package into the model in real time for real-time prediction; when the model predicts that the package is at risk of being lost, issue an early warning. Use SMS, email, mobile application push, etc. to send early warning information to express management personnel and recipients. The early warning information may include information such as the location, status, estimated arrival time, and possible loss risks of the package; express management personnel can take corresponding measures based on the early warning information, such as contacting the courier to verify the location and status of the package, adjusting the transportation route, etc. At the same time, the early warning information can be tracked and recorded for subsequent analysis and improvement of the risk early warning model.

[0039] 204. Analyze historical data using a loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimize transportation routes based on the predicted high-risk links and factors for package loss that meet preset requirements; In this embodiment, the use of the loss risk prediction model to analyze historical data, predict high-risk links and factors for package loss that meet preset requirements, and optimize the transportation route based on the predicted high-risk links and factors for package loss that meet preset requirements includes: Build a missing risk prediction model based on machine learning algorithms; Construct training and test sets for training loss risk prediction models based on historical data; Using the training set to train the loss risk prediction model, to obtain a trained loss risk prediction model; The trained loss risk prediction model is evaluated using the test set, and the current loss risk prediction model is further optimized according to the evaluation results to obtain an optimized loss risk prediction model; Based on the pre-processed parcel data of each logistics link, the parcel loss rate of different links is analyzed using the optimized loss risk prediction model. When the parcel loss rate is higher than the preset value, the current link is considered to be a high-risk link; For high-risk links, the transportation route optimization model is used to optimize the transportation route; specifically, based on the location information, destination, transportation time and transportation cost of the package as the optimization target, the transportation route optimization model is constructed with road conditions and traffic restrictions as constraints; the constructed transportation route optimization model is optimized and solved using a genetic algorithm to obtain the optimal transportation route.

[0040] This embodiment designs a closed-loop system from data collection, analysis, prediction, optimization to feedback, and continuously iterates and improves the solution, thereby effectively reducing the occurrence of express loss incidents and improving overall work efficiency and customer satisfaction.

[0041] See also Figure 3 The third embodiment of the express delivery loss optimization method in the embodiment of the present invention includes: 301. Obtaining parcel data of each logistics link, and preprocessing the obtained parcel data of each logistics link; 302. Based on the pre-processed parcel data of each logistics link, the express business data model is constructed according to the entity relationship model; 303. Analyze the risk factors of package loss based on the express delivery business data model; 304. Based on the pre-processed parcel data of each logistics link, according to the risk factors of parcel loss, the risk of parcel loss is predicted using the risk warning model; 305. Analyze historical data using a loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimize transportation routes based on the predicted high-risk links and factors for package loss that meet preset requirements; In this embodiment, historical express business data, including the location information, status information, transportation route, etc. of the package, is collected and sorted, and the data is divided into a training set and a test set for model training and evaluation; a suitable machine learning algorithm is selected, such as a decision tree, random forest, support vector machine, neural network, etc. According to the characteristics of the data and the needs of the problem, a suitable algorithm is selected for analysis and prediction; the data is preprocessed, including data cleaning, feature extraction, data standardization, etc., and data cleaning tools or scripts are used to remove noise and outliers in the data; feature extraction algorithms such as principal component analysis (PCA) and linear discriminant analysis (LDA) can be used to extract features in the data; data standardization methods such as Z-score standardization and Min-Max standardization are used to standardize the data to the same scale; the machine learning algorithm is trained using the training set to establish a parcel loss risk prediction model. By continuously adjusting the parameters of the model, the accuracy and generalization ability of the model are improved; the trained model is evaluated and optimized, and the model is evaluated using the test set to calculate the model's accuracy, recall rate, F1 value and other indicators. According to the evaluation results, adjust the model parameters or select other algorithms to further optimize the model performance; use the trained package loss risk prediction model to analyze historical data, predict the high-risk links and factors of package loss, analyze the package loss rate in different links, and find out the high-risk links; analyze the impact of different factors on the risk of package loss and find out the high-risk factors; visualize and analyze the prediction results to better understand the risk of package loss, and use data visualization tools such as Tableau, PowerBI, etc. to display the prediction results in the form of charts to facilitate express management personnel to analyze and make decisions.

[0042] Establish a transportation route optimization model, take the location information, destination, transportation time, transportation cost and other factors of the package as optimization targets, and take road conditions, traffic restrictions and other factors as constraints; use the optimization algorithm to solve the transportation route optimization model, find the optimal transportation route, use the crossover and mutation operations of the genetic algorithm to continuously evolve the population and find the optimal solution; use the speed update and position update formulas of the particle swarm optimization algorithm to continuously adjust the position of the particles and find the optimal solution; use the temperature drop and probability acceptance mechanism of the simulated annealing algorithm to continuously search the solution space and find the optimal solution; evaluate and verify the optimized transportation route, use actual express business data to test the optimized transportation route, calculate indicators such as transportation time and transportation cost, and evaluate the optimization effect. If the optimization effect is not ideal, you can adjust the parameters of the optimization algorithm or choose other algorithms to further optimize the transportation route.

[0043] 306. Deploy high-definition monitoring equipment at express delivery outlets and use video analysis technology to record and automatically monitor behaviors such as violent sorting and non-compliant safety inspections to reduce cargo losses.

[0044] In this embodiment, the OpenCV library and Haar cascade classifier are used for face recognition to enhance the security of express delivery sites; face recognition technology can be used for identity authentication and access control to ensure that only authorized personnel can enter specific areas. Based on AI artificial intelligence algorithms, through machine vision, image recognition and other technologies, intelligent analysis of video surveillance data, such as fireworks recognition, license plate recognition, face recognition, etc., real-time analysis of vehicles, goods, people and site dynamics, effective monitoring of the on-site conditions of each express delivery site, timely detection of safety hazards, help managers make timely decisions, and improve the early warning capabilities of safety risks of express delivery sites.

[0045] The above describes the express delivery loss optimization method in the embodiment of the present invention. The following describes the express delivery loss optimization device in the embodiment of the present invention. Figure 4 , an embodiment of the express delivery loss optimization device in the embodiment of the present invention includes: The data acquisition module 401 is used to acquire the package data of each logistics link and pre-process the acquired package data of each logistics link; In this embodiment, the data acquisition module 401 includes: Obtain package data including collection, sorting, transportation and delivery through API interface; The temperature sensor, humidity sensor and vibration sensor installed on the package collect the status information of the package in real time; Get the size, weight and destination information of the package by scanning the barcode on the courier; The obtained package data of each logistics link is cleaned, feature extracted and standardized to obtain the pre-processed package data of each logistics link.

[0046] A business data model building module 402 is used to build an express business data model according to an entity relationship model based on the pre-processed package data of each logistics link; In this embodiment, the business data model building module 402 includes: Based on the pre-processed parcel data of each logistics link, the entity relationship model is used to associate the information of collection, sorting, transportation, delivery and other links to form a complete express business data model.

[0047] The risk factor analysis module 403 is used to analyze the risk factors of package loss based on the express delivery business data model; In this embodiment, the risk factor analysis module 403 includes: Based on the express delivery business data model, the risk factors of package loss are analyzed through methods including data visualization, association rule mining and cluster analysis.

[0048] The risk warning module 404 is used to predict the risk of package loss using a risk warning model based on the pre-processed package data of each logistics link and the risk factors of package loss.

[0049] In this embodiment, the risk warning module 404 includes: Build risk warning models based on machine learning algorithms; Constructing a training set and a test set for training a risk warning model; specifically, obtaining parcel data of historical logistics links, and preprocessing the obtained parcel data of historical logistics links, including: data cleaning, feature extraction, and data standardization; dividing the preprocessed parcel data of historical logistics links into a training set and a test set; when the training set and / or the test set is an unbalanced data set, performing mixed sampling on the unbalanced data set to obtain a balanced data set; wherein, performing mixed sampling on the unbalanced data set includes: using an oversampling method to increase the number of minority class samples; and using an undersampling method to reduce the number of majority class samples.

[0050] The risk warning model is trained using the training set to obtain a trained risk warning model; Use the test set to evaluate the trained risk warning model, and further optimize the current risk warning model based on the evaluation results to obtain an optimized risk warning model; Using the optimized risk warning model, based on the pre-processed package data of each logistics link, it is predicted whether the package is at risk of loss. When there is a risk of loss, an early warning is triggered.

[0051] The transportation route optimization module 405 is used to analyze the historical data using the loss risk prediction model, predict the high-risk links and factors of the package loss that meet the preset requirements, and optimize the transportation route based on the predicted high-risk links and factors of the package loss that meet the preset requirements; In this embodiment, the transportation route optimization module 405 includes: Build a missing risk prediction model based on machine learning algorithms; Construct training and test sets for training loss risk prediction models based on historical data; Using the training set to train the loss risk prediction model, to obtain a trained loss risk prediction model; The trained loss risk prediction model is evaluated using the test set, and the current loss risk prediction model is further optimized according to the evaluation results to obtain an optimized loss risk prediction model; Based on the pre-processed parcel data of each logistics link, the parcel loss rate of different links is analyzed using the optimized loss risk prediction model. When the parcel loss rate is higher than the preset value, the current link is considered to be a high-risk link; For high-risk links, the transportation route optimization model is used to optimize the transportation route; specifically, based on the location information, destination, transportation time and transportation cost of the package as the optimization target, the transportation route optimization model is constructed with road conditions and traffic restrictions as constraints; the constructed transportation route optimization model is optimized and solved using a genetic algorithm to obtain the optimal transportation route.

[0052] above Figure 4 The express loss optimization device in the embodiment of the present invention is described in detail from the perspective of modular functional entities, and the electronic device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0053] Figure 57 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 700 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 710 (for example, one or more processors) and a memory 720, and one or more storage media 730 (for example, one or more mass storage devices) storing application programs 733 or data 732. Among them, the memory 720 and the storage medium 730 can be temporary storage or permanent storage. The program stored in the storage medium 730 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the electronic device 700. Furthermore, the processor 710 may be configured to communicate with the storage medium 730 to execute a series of instruction operations in the storage medium 730 on the electronic device 700.

[0054] The electronic device 700 may also include one or more power supplies 740, one or more wired or wireless network interfaces 750, one or more input and output interfaces 750, and / or one or more operating systems 731, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 5 The structure of the electronic device shown does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0055] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the express loss optimization method.

[0056] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0057] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0058] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing lost express delivery, characterized in that: include: Obtain the parcel data of each logistics link, and pre-process the obtained parcel data of each logistics link; Based on the pre-processed parcel data of each logistics link, the express business data model is constructed according to the entity relationship model; Analyze the risk factors of package loss based on the express delivery business data model; Based on the pre-processed parcel data of each logistics link and the risk factors of parcel loss, the risk of parcel loss is predicted using the risk warning model.

2. The express delivery loss optimization method according to claim 1, characterized in that: The obtaining of parcel data of each logistics link includes: Obtain package data including collection, sorting, transportation and delivery through API interface; The temperature sensor, humidity sensor and vibration sensor installed on the package collect the status information of the package in real time; Get the size, weight and destination information of the package by scanning the barcode on the courier; The obtained package data of each logistics link is cleaned, feature extracted and standardized to obtain the pre-processed package data of each logistics link.

3. The express delivery loss optimization method according to claim 1, characterized in that: The risk factors for parcel loss analyzed based on the express delivery business data model include: Based on the express delivery business data model, the risk factors of package loss are analyzed through methods including data visualization, association rule mining and cluster analysis.

4. The express delivery loss optimization method according to claim 1, characterized in that: The risk of package loss is predicted by using a risk warning model based on the pre-processed package data of each logistics link according to the risk factors of package loss, including: Build risk warning models based on machine learning algorithms; Construct training sets and test sets for training risk warning models; The risk warning model is trained using the training set to obtain a trained risk warning model; Use the test set to evaluate the trained risk warning model, and further optimize the current risk warning model based on the evaluation results to obtain an optimized risk warning model; Using the optimized risk warning model, based on the pre-processed package data of each logistics link, it is predicted whether the package is at risk of loss. When there is a risk of loss, an early warning is triggered.

5. The express delivery loss optimization method according to claim 4, characterized in that: The training set and test set for constructing the training risk warning model include: Obtain the parcel data of historical logistics links, and pre-process the parcel data of historical logistics links, including: data cleaning, feature extraction and data standardization; Divide the pre-processed historical logistics parcel data into training set and test set; When the training set and / or the test set is an unbalanced data set, the unbalanced data set is mixed and sampled to obtain a balanced data set; The mixed sampling of the imbalanced data set includes: using an oversampling method to increase the number of minority class samples; and using an undersampling method to reduce the number of majority class samples.

6. The express delivery loss optimization method according to claim 1, characterized in that: The method further includes: analyzing historical data using a loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimizing transportation routes based on the predicted high-risk links and factors for package loss that meet preset requirements; The method of analyzing historical data using the loss risk prediction model to predict high-risk links and factors for package loss that meet preset requirements, and optimizing the transportation route based on the predicted high-risk links and factors for package loss that meet preset requirements, includes: Build a missing risk prediction model based on machine learning algorithms; Construct training and test sets for training loss risk prediction models based on historical data; Using the training set to train the loss risk prediction model, to obtain a trained loss risk prediction model; The trained loss risk prediction model is evaluated using the test set, and the current loss risk prediction model is further optimized according to the evaluation results to obtain an optimized loss risk prediction model; Based on the pre-processed parcel data of each logistics link, the parcel loss rate of different links is analyzed using the optimized loss risk prediction model. When the parcel loss rate is higher than the preset value, the current link is considered to be a high-risk link; For high-risk links, transportation routes are optimized using transportation route optimization models.

7. The express delivery loss optimization method according to claim 6, characterized in that: The above-mentioned method of optimizing the transportation route by using the transportation route optimization model for high-risk links includes: Based on the location information, destination, transportation time and transportation cost of the package as optimization objectives, and including road conditions and traffic restrictions as constraints, a transportation route optimization model is constructed; Genetic algorithms are used to optimize and solve the constructed transportation route optimization model to obtain the optimal transportation route.

8. A lost express optimization device, characterized in that: include: The data acquisition module is used to acquire the package data of each logistics link and pre-process the acquired package data of each logistics link; The business data model building module is used to build the express business data model based on the pre-processed parcel data of each logistics link according to the entity relationship model; The risk factor analysis module is used to analyze the risk factors of package loss based on the express delivery business data model; The risk warning module is used to predict the risk of package loss based on the pre-processed package data of each logistics link and the risk factors of package loss using the risk warning model.

9. An electronic device, comprising a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the electronic device executes each step of the express loss optimization method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the various steps of the express loss optimization method as described in any one of claims 1-7 are implemented.

Images