Intelligent expediting method, device, equipment and storage medium for logistics orders

Abstract

The present invention relates to the field of information processing technology, and in particular to an intelligent expediting method, device, equipment and storage medium for logistics orders. Standardized related data are predicted through a delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, and the salespersons and business pickup and dispatch areas are sorted according to the predicted value of the pickup and dispatch delay rate. The salespersons and business pickup and dispatch areas are screened according to the sorting result according to a preset ratio to obtain information on salespersons and business pickup and dispatch areas with high delay rates, and the information on salespersons and business pickup and dispatch areas with high delay rates is analyzed to obtain a delay cause analysis result. The information on salespersons and business pickup and dispatch areas with high delay rates and the delay cause analysis result are input into a path planning model for path planning to obtain pickup and dispatch path planning information, which can optimize the pickup and dispatch paths of salespersons and business pickup and dispatch areas with high delay rates, prevent pickup and dispatch delays, and improve user satisfaction.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to an intelligent expediting method, device, equipment and storage medium for logistics orders. Background Art

[0002] At present, the logistics service system of e-commerce platforms can automatically generate collection and dispatch routes, provide route guidance for salesmen, and accurately predict and notify the e-commerce platform and users of the specific time point of the next logistics track update, greatly improving the timeliness and predictability of logistics information. However, in the face of growing logistics volume and complex and changing transportation environment, traditional prediction models are unable to adjust existing collection and dispatch routes when faced with salesmen and business collection and dispatch areas with high delay rates, and are unable to dynamically optimize the collection and dispatch routes, making it difficult to urge salesmen to deliver, resulting in collection and dispatch delays. It may also cause the platform to be unable to promptly and accurately inform users of logistics dynamics, thereby affecting user experience and the reputation of e-commerce platforms. Summary of the invention

[0003] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide an intelligent expediting method, device, equipment and storage medium for logistics orders that can automatically identify salesmen and business collection and delivery areas with high delay rates, and optimize the collection and delivery paths of salesmen and business collection and delivery areas with high delay rates, prevent collection and delivery delays, and improve user satisfaction.

[0004] The first aspect of the present invention provides an intelligent expediting method for logistics orders, comprising: extracting relevant data of all salesmen and business collection and dispatch areas from a business system database, performing data preprocessing on the relevant data to obtain standardized relevant data; using a preset delay prediction model to predict the standardized relevant data to obtain a collection and dispatch delay rate prediction value for each of the salesmen and business collection and dispatch areas, sorting the salesmen and business collection and dispatch areas according to the collection and dispatch delay rate prediction value to obtain a sorting result; screening the salesmen and business collection and dispatch areas according to the sorting result according to a preset ratio to obtain information on salesmen and business collection and dispatch areas with high delay rates, analyzing the information on salesmen and business collection and dispatch areas with high delay rates to obtain a delay cause analysis result; using a genetic algorithm or an ant colony optimization algorithm to construct a path planning model, inputting the information on salesmen and business collection and dispatch areas with high delay rates and the delay cause analysis result into the path planning model for path planning to obtain collection and dispatch path planning information; uploading the collection and dispatch path planning information to a blockchain, and sending the collection and dispatch path planning information to a salesman terminal.

[0005] Optionally, in a first implementation method of the first aspect of the present invention, the relevant data of all salesmen and business dispatch areas are extracted from the business system database, and the relevant data are preprocessed to obtain standardized relevant data, including: extracting the relevant data of all salesmen and business dispatch areas from the business system database, the relevant data including salesman information, business dispatch area information and order information; performing data cleansing on the salesman information, the business dispatch area information and the order information to obtain cleaned relevant data; correcting abnormal values ​​of the cleaned relevant data to obtain corrected relevant data, and filling missing values ​​of the corrected relevant data to obtain filled relevant data; based on a preset target standard format, performing format conversion on the filled relevant data to obtain standardized relevant data.

[0006] Optionally, in a second implementation method of the first aspect of the present invention, the preset delay prediction model is used to predict the standardized related data to obtain a predicted value of the pickup and dispatch delay rate for each of the salespersons and business pickup and dispatch areas, and the salespersons and business pickup and dispatch areas are sorted according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result, including: obtaining historical salesperson information, historical business pickup and dispatch area information and historical order information; integrating the historical salesperson information, historical business pickup and dispatch area information and historical order information to obtain historical related data; dividing the historical related data according to a preset ratio to obtain a training set, a validation set and a test set; selecting a recurrent neural network model, and training the recurrent neural network model with the training set to obtain a delay prediction model; tuning and evaluating the delay prediction model with the validation set and the test set respectively; using the delay prediction model to predict the standardized related data to obtain a predicted value of the pickup and dispatch delay rate for each of the salespersons and business pickup and dispatch areas; sorting the salespersons and business pickup and dispatch areas according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result.

[0007] Optionally, in a third implementation method of the first aspect of the present invention, the salesmen and business dispatch areas are screened according to a preset ratio based on the sorting results to obtain information on salesmen and business dispatch areas with high delay rates, and the information on salesmen and business dispatch areas with high delay rates is analyzed to obtain delay cause analysis results, including: screening the salesmen and business dispatch areas according to a preset ratio based on the sorting results to obtain information on salesmen and business dispatch areas with high delay rates; identifying unfinished order information in the information on salesmen and business dispatch areas with high delay rates; determining whether the dispatch time of the unfinished order information exceeds a preset time threshold; if so, analyzing the information on salesmen and business dispatch areas with high delay rates to obtain delay cause analysis results.

[0008] Optionally, in a fourth implementation method of the first aspect of the present invention, a path planning model is constructed using a genetic algorithm or an ant colony optimization algorithm, and the information of the salesperson with a high delay rate and the business pick-up and dispatch area and the delay cause analysis results are input into the path planning model for path planning to obtain the pick-up and dispatch path planning information, including: constructing a path planning model using a genetic algorithm or an ant colony optimization algorithm; inputting the information of the salesperson with a high delay rate and the business pick-up and dispatch area and the delay cause analysis results into the path planning model for path planning to obtain a specific pick-up and dispatch driving path, a pick-up and dispatch order, and an estimated pick-up and dispatch time; integrating the specific pick-up and dispatch driving path, the pick-up and dispatch order, and the estimated pick-up and dispatch time into a data structure to obtain the pick-up and dispatch path planning information.

[0009] Optionally, in a fifth implementation method of the first aspect of the present invention, uploading the pickup and dispatch path planning information to the blockchain, and sending the pickup and dispatch path planning information to the salesperson terminal, includes: generating a visual pickup and dispatch path planning chart based on the pickup and dispatch path planning information; sending the pickup and dispatch path planning chart to the salesperson terminal; encrypting the pickup and dispatch path planning chart to obtain an encrypted pickup and dispatch path planning chart, and uploading the encrypted pickup and dispatch path planning chart to the blockchain.

[0010] Optionally, in a sixth implementation method of the first aspect of the present invention, after uploading the pickup and dispatch path planning information to the blockchain and sending the pickup and dispatch path planning information to the salesperson terminal, it also includes: creating a listener, and using the listener to listen to the completion status of the pickup and dispatch path planning information; when the listener listens to the completion status, obtaining the salesperson's pickup and dispatch completion information corresponding to the current completion status; evaluating the path optimization effect of the salesperson's pickup and dispatch completion information to obtain an evaluation score; judging whether the evaluation score reaches a preset evaluation threshold; if not, generating a model adjustment parameter; using the model adjustment parameter to iteratively update the path planning model; when the listener fails to listen to the completion status within a preset time, generating an early warning message, and sending the early warning message to the management terminal.

[0011] The second aspect of the present invention provides an intelligent expediting device for logistics orders, including: an extraction and processing module, which is used to extract relevant data of all salesmen and business collection and dispatch areas from a business system database, perform data preprocessing on the relevant data, and obtain standardized relevant data; a prediction and sorting module, which is used to predict the standardized relevant data using a preset delay prediction model, obtain a collection and dispatch delay rate prediction value for each of the salesmen and business collection and dispatch areas, and sort the salesmen and business collection and dispatch areas according to the collection and dispatch delay rate prediction value to obtain a sorting result; a screening and analysis module, which is used to sort the salesmen and business collection and dispatch areas according to the sorting result. The business dispatch area is screened according to a preset ratio to obtain the salesperson with high delay rate and the business dispatch area information, and the salesperson with high delay rate and the business dispatch area information are analyzed to obtain the delay cause analysis result; a planning module is constructed to construct a path planning model using a genetic algorithm or an ant colony optimization algorithm, and the salesperson with high delay rate and the business dispatch area information and the delay cause analysis result are input into the path planning model for path planning to obtain the dispatch path planning information; an upload and sending module is used to upload the dispatch path planning information to the blockchain, and send the dispatch path planning information to the salesperson terminal.

[0012] Optionally, in a first implementation method of the second aspect of the present invention, the extraction and processing module includes: an extraction unit, used to extract relevant data of all salesmen and business dispatch areas from a business system database, the relevant data including salesman information, business dispatch area information and order information; a cleaning unit, used to clean the salesman information, the business dispatch area information and the order information to obtain cleaned relevant data; a correction and filling unit, used to correct abnormal values ​​of the cleaned relevant data to obtain corrected relevant data, and fill missing values ​​of the corrected relevant data to obtain filled relevant data; a conversion unit, used to convert the format of the filled relevant data based on a preset target standard format to obtain standardized relevant data.

[0013] Optionally, in a second implementation method of the second aspect of the present invention, the prediction and sorting module includes: an acquisition unit for acquiring historical salesperson information, historical business pick-up and dispatch area information and historical order information; a first integration unit for integrating the historical salesperson information, historical business pick-up and dispatch area information and historical order information to obtain historical related data; a division unit for dividing the historical related data according to a preset ratio to obtain a training set, a validation set and a test set; a training selection unit for selecting a recurrent neural network model, training the recurrent neural network model with the training set to obtain a delay prediction model; a tuning and evaluation unit for tuning and evaluating the delay prediction model with the validation set and the test set respectively; a prediction unit for using the delay prediction model to predict the standardized related data to obtain a pick-up and dispatch delay rate prediction value for each of the salespersons and business pick-up and dispatch areas; a sorting unit for sorting the salespersons and business pick-up and dispatch areas according to the pick-up and dispatch delay rate prediction value to obtain a sorting result.

[0014] Optionally, in a third implementation of the second aspect of the present invention, the screening and analysis module includes: a screening unit, used to screen the salesmen and business dispatch areas according to a preset ratio based on the sorting results, and obtain the information of the salesmen and business dispatch areas with high delay rates; an identification unit, used to identify the unfinished order information in the information of the salesmen and business dispatch areas with high delay rates; a judgment unit, used to judge whether the dispatch time of the unfinished order information exceeds a preset time threshold; and an analysis unit, used to analyze the information of the salesmen and business dispatch areas with high delay rates, and obtain the delay cause analysis results if so.

[0015] Optionally, in a fourth implementation method of the second aspect of the present invention, the planning module includes: a construction unit, which is used to construct a path planning model using a genetic algorithm or an ant colony optimization algorithm; a planning unit, which is used to input the information of the salesperson with high delay rate and the business pick-up and dispatch area and the delay cause analysis results into the path planning model for path planning, and obtain a specific pick-up and dispatch route, pick-up and dispatch sequence, and an estimated pick-up and dispatch time; a second integration unit, which is used to integrate the specific pick-up and dispatch route, the pick-up and dispatch sequence, and the estimated pick-up and dispatch time into a data structure to obtain the pick-up and dispatch route planning information.

[0016] Optionally, in a fifth implementation of the second aspect of the present invention, the upload and sending module includes: a generation unit, used to generate a visual pickup and dispatch path planning chart based on the pickup and dispatch path planning information; a sending unit, used to send the pickup and dispatch path planning chart to the salesperson terminal; an encryption upload unit, used to encrypt the pickup and dispatch path planning chart to obtain an encrypted pickup and dispatch path planning chart, and upload the encrypted pickup and dispatch path planning chart to the blockchain.

[0017] Optionally, in a sixth implementation method of the second aspect of the present invention, it also includes: creating a monitoring module, used to create a listener, and using the listener to monitor the completion status of the pick-up and dispatch path planning information; an acquisition module, used to obtain the salesperson's pick-up and dispatch completion information corresponding to the current completion status when the listener monitors the completion status; an evaluation module, used to evaluate the path optimization effect of the salesperson's pick-up and dispatch completion information to obtain an evaluation score; a judgment module, used to judge whether the evaluation score reaches a preset evaluation threshold; a generation module, used to generate model adjustment parameters if not; an update module, used to iteratively update the path planning model using the model adjustment parameters; a generation and sending module, used to generate an early warning message when the listener fails to monitor the completion status within a preset time, and send the early warning message to the management terminal.

[0018] The third aspect of the present invention provides an intelligent dispatching device for logistics orders, which includes: a memory and at least one processor, wherein instructions are stored in the memory; at least one of the processors calls the instructions in the memory so that the intelligent dispatching device for logistics orders executes each step of the intelligent dispatching method for logistics orders described in any one of the above items.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed by a processor, implement the various steps of any of the above-mentioned intelligent dispatching methods for logistics orders.

[0020] In the technical solution of the present invention, standardized related data is predicted by a delay prediction model to obtain a prediction value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, the salespersons and business pickup and dispatch areas are sorted according to the prediction value of the pickup and dispatch delay rate, and the salespersons and business pickup and dispatch areas are screened according to the sorting result according to a preset ratio to obtain information on salespersons and business pickup and dispatch areas with high delay rates, and the information on salespersons and business pickup and dispatch areas with high delay rates is analyzed to obtain a delay cause analysis result, and the information on salespersons and business pickup and dispatch areas with high delay rates and the delay cause analysis result are input into a path planning model for path planning to obtain pickup and dispatch path planning information, which can optimize the pickup and dispatch paths for salespersons and business pickup and dispatch areas with high delay rates to prevent pickup and dispatch delays and improve user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A first flow chart of the intelligent expediting method for logistics orders provided by an embodiment of the present invention;

[0022] Figure 2 A second flow chart of the intelligent expediting method for logistics orders provided in an embodiment of the present invention;

[0023] Figure 3 A third flow chart of the intelligent expediting method for logistics orders provided in an embodiment of the present invention;

[0024] Figure 4 A fourth flow chart of the intelligent expediting method for logistics orders provided in an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of the structure of an intelligent expediting device for logistics orders provided in an embodiment of the present invention;

[0026] Figure 6 Another structural schematic diagram of the intelligent expediting device for logistics orders provided in an embodiment of the present invention;

[0027] Figure 7 A schematic diagram of the structure of an intelligent expediting device for logistics orders provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The present invention provides an intelligent expediting method, device, equipment and storage medium for logistics orders, which can automatically identify salesmen and business collection and dispatch areas with high delay rates, and can optimize the collection and dispatch paths of salesmen and business collection and dispatch areas with high delay rates, prevent collection and dispatch delays, and improve user satisfaction.

[0029] 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.

[0030] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 An embodiment of the intelligent expediting method for logistics orders in the embodiment of the present invention includes:

[0031] 101. Extract relevant data of all salesmen and business dispatch areas from the business system database, pre-process the relevant data, and obtain standardized relevant data;

[0032] In this embodiment, relevant data of all salesmen and business dispatch areas are extracted from the business system database, duplicate data is removed, such as duplicate records of the same salesman, missing values ​​are processed, common methods include filling, such as mean filling, mode filling, or deleting missing data, checking outliers and erroneous data, ensuring data consistency and accuracy, unifying the formats of data from different sources, such as unifying the time and date format, converting the numerical type, etc., converting text data into numerical format, for example, converting categorical variables into numerical encoding or one-hot encoding, and standardizing numerical features. Common methods are Z-score standardization or Min-Max normalization. This can eliminate the influence of different feature dimensions and ensure that each feature contributes to the result in a balanced manner during model training. Discrete data such as the salesman's work area is uniformly encoded to ensure consistency of model input.

[0033] 102. Use a preset delay prediction model to predict the standardized related data to obtain a predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, and sort the salespersons and business pickup and dispatch areas according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result;

[0034] In this embodiment, the standardized data set will include the historical collection and dispatch records of each salesperson, the collection and dispatch area where the salesperson is located, and other factors that may affect the collection and dispatch delay. Features that have a significant impact on delay prediction are selected from the standardized data as model input, the previously trained delay prediction model is loaded, and the standardized data is input into the model for prediction. The collection and dispatch delay rate prediction value of each salesperson and collection and dispatch area. For each salesperson and collection and dispatch area, the model will generate a delay rate prediction value, indicating the possible delays that may occur in the salesperson or area in future collection and dispatch tasks, and the salespersons are ranked according to their delay rate prediction values.

[0035] 103. According to the sorting results, the salespersons and the business dispatching areas are screened according to a preset ratio to obtain the information of the salespersons and the business dispatching areas with high delay rates, and the information of the salespersons and the business dispatching areas with high delay rates is analyzed to obtain the delay cause analysis results;

[0036] In this embodiment, a screening ratio is preset according to business needs or risk control strategies. This ratio usually reflects the high-risk range that the company wants to focus on. For example, if the top 10% of salesmen and regions with the highest delay rates are selected, then the 10% with higher delay rates in the sorting results will be screened out, and the screened salesmen and regions with high delay rates will be deeply analyzed to identify the potential causes of their delays. The causes of delays for salesmen include workload and task allocation, geographical factors, external environmental influences, and personal abilities and experience of salesmen. The causes of delays for business dispatch areas include regional workload, traffic and infrastructure issues, regional resource scheduling issues, and external environmental factors. After completing the delay cause analysis, a detailed cause analysis report needs to be provided for each high-delay salesman and region.

[0037] 104. Use a genetic algorithm or an ant colony optimization algorithm to build a path planning model, input the information of the salesperson with high delay rate and the business pick-up and dispatch area and the delay cause analysis results into the path planning model to perform path planning, and obtain the pick-up and dispatch path planning information;

[0038] In this embodiment, a genetic algorithm or an ant colony optimization algorithm is used to construct a path planning model, and the information of salesmen with high delay rates, business pick-up and dispatch areas, and delay cause analysis results are input into the path planning model for path planning. After iterations of the genetic algorithm or the ant colony optimization algorithm, the pick-up and dispatch path planning information is obtained.

[0039] 105. Upload the route planning information of the pickup and dispatch to the blockchain, and send the route planning information of the pickup and dispatch to the salesperson terminal;

[0040] In this embodiment, a suitable blockchain platform is selected, and data is uploaded on the blockchain for automated operations with the help of smart contracts. When uploading path planning information, the information is encrypted using digital signature technology to ensure that the source of the information is traceable and prevent information tampering, and the path planning information is pushed to the salesperson's terminal.

[0041] In an embodiment of the present invention, standardized related data is predicted by a delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, the salespersons and business pickup and dispatch areas are sorted according to the predicted value of the pickup and dispatch delay rate, and the salespersons and business pickup and dispatch areas are screened according to the sorting result according to a preset ratio to obtain information on salespersons and business pickup and dispatch areas with high delay rates, the information on salespersons and business pickup and dispatch areas with high delay rates is analyzed to obtain a delay cause analysis result, the information on salespersons and business pickup and dispatch areas with high delay rates and the delay cause analysis result are input into a path planning model for path planning to obtain pickup and dispatch path planning information, which can optimize the pickup and dispatch paths for salespersons and business pickup and dispatch areas with high delay rates, prevent pickup and dispatch delays, and improve user satisfaction.

[0042] See also Figure 2 The second embodiment of the intelligent expediting method for logistics orders in the embodiment of the present invention includes:

[0043] 201. Extract relevant data of all salesmen and business delivery areas from the business system database, the relevant data including salesman information, business delivery area information and order information;

[0044] In this embodiment, the business system database includes multiple tables, each table stores different types of information. The salesperson information table stores the personal information of all salespeople, such as name, contact information, work number, department, etc. The business delivery area table records the delivery area information for each salesperson, including area ID, area name, coverage, etc. The order information table stores all order data related to the salesperson, which may include order number, customer information, order time, order status, order delivery address, etc.

[0045] 202. Clean the salesperson information, business delivery area information and order information to obtain relevant data after cleaning;

[0046] In this embodiment, duplicate data of salesperson information, business dispatch area information, and order information are deleted to obtain cleaned related data.

[0047] 203. Correcting abnormal values ​​of the cleaned relevant data to obtain corrected relevant data, and filling missing values ​​of the corrected relevant data to obtain filled relevant data;

[0048] In this embodiment, a reasonable range is defined according to the type of data and the business background. A boxplot method can be used to check the data distribution and identify outliers. For numerical outliers, they can be corrected by setting a reasonable upper or lower limit. For time fields, if the order date is a future date, it can be corrected to the current date. For categorical data errors, if the area name is an invalid value, it can be corrected to a default value (such as "unknown area") or corrected through a mapping table. After correction, corrected related data is obtained, and the location of missing values ​​in each field is detected. For numerical fields, the mean or median of the field can be used to fill the missing value, which is particularly suitable for data without obvious trends. In some cases, the missing value can be filled with 0 (such as the order amount), and the adjacent data before and after the missing value can be used to fill it, which is particularly suitable for time series data. For categorical data, the most frequently occurring value in the field can be selected for filling, such as some fields (such as area name, customer address) can be filled with "unknown" or other preset values. If the area ID of an order is missing, the area ID can be inferred based on the salesperson ID of the order to obtain the filled related data.

[0049] 204. Based on the preset target standard format, convert the format of the filled relevant data to obtain standardized relevant data;

[0050] In this embodiment, a target standard format is determined and the data is converted into this standard format. The data format conversion steps include field name standardization, data type conversion, date and time format unification, unit standardization, and filling in missing value markers.

[0051] 205. Obtain historical salesperson information, historical business delivery area information and historical order information;

[0052] In this embodiment, historical salesperson information including basic information, position, work area, etc. of the salesperson is obtained, historical business dispatch area information is obtained, and the business dispatch area information is used to analyze the market coverage of the salesperson, and historical order information is obtained. The historical order information records the specific information of the products or services sold by the salesperson, including details such as order generation, payment, and delivery.

[0053] 206. Integrate historical salesperson information, historical business delivery area information and historical order information to obtain historical related data;

[0054] In this embodiment, the historical salesperson information includes the basic information of the salesperson, the historical business dispatch area information includes the regional allocation corresponding to the salesperson, the historical order information records the orders and sales for which the salesperson is responsible, and clarifies which fields the integrated data set contains. The data tables of the salesperson, region, and order are associated and connected. After the historical salesperson information, historical business dispatch area information, and historical order information are integrated, an integrated table or data set containing historical related data is obtained.

[0055] 207. Divide the historical related data according to a preset ratio to obtain a training set, a validation set, and a test set;

[0056] In this embodiment, it is necessary to determine the ratio of the training set, the validation set, and the test set, integrate the historical salesperson information, the historical business delivery area information, and the historical order information into a single data set, randomly shuffle the integrated data set, and divide it according to the preset ratio to obtain the training set, the validation set, and the test set.

[0057] 208. Select a recurrent neural network model, and train the recurrent neural network model using the training set to obtain a delay prediction model;

[0058] In this embodiment, a recurrent neural network model is selected, the data is converted into a format suitable for input into the RNN, the model is trained using the training set data, and the training process is monitored by drawing training and validation loss curves to determine whether overfitting occurs.

[0059] 209. Tune and evaluate the delay prediction model using the validation set and test set respectively;

[0060] In this embodiment, the delay prediction model is tuned through the validation set, and the optimal hyperparameters are selected to ensure that the model can achieve the best performance during the training process. After the model tuning is completed, the test set is used to evaluate the final performance of the model, and the model architecture is continuously adjusted according to the results of the validation set and the test set to improve the performance.

[0061] 210. Use the delay prediction model to predict the standardized related data to obtain the predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area;

[0062] In this embodiment, the delay prediction model is used to predict the standardized related data to obtain the predicted value of the delay rate of each salesperson and business dispatch area. For each salesperson and business dispatch area, the predicted delay rate can be evaluated individually to identify which areas or salespersons have larger prediction errors.

[0063] 211. Sort the salesmen and the business delivery areas according to the predicted value of the delivery delay rate to obtain the sorting result;

[0064] In this embodiment, once the predicted values ​​are obtained, the salesmen and business dispatch areas can be sorted according to these predicted values, and the predicted values ​​can be bound to the relevant salesman information (such as salesman ID or area ID) to obtain the sorting results.

[0065] 212. According to the sorting results, the salesmen and the business dispatching areas are screened according to the preset ratio to obtain the salesmen and business dispatching areas with high delay rates;

[0066] In this embodiment, the screening ratio refers to the ratio of salesmen or regions with high delay rates that you want to extract from the sorting results. For example, if you choose to screen out 10% of salesmen or regions with high delay rates, then the 10% of salesmen or regions with the highest delay rates will be selected from the sorted list, and the top N salesmen or regions with the lowest ranking will be selected through the index.

[0067] 213. Identify the unfinished order information in the salespersons with high delay rate and the business pick-up and dispatch area information;

[0068] In this embodiment, unfinished orders are filtered out from salesmen and regions with high delay rates. The filtering function can be used to extract orders with high delay rates and an order status of "unfinished", thus obtaining unfinished orders with high delay rates.

[0069] 214. Determine whether the delivery time of the uncompleted order information exceeds the preset time threshold;

[0070] In this embodiment, the collection and dispatch time is calculated. For each order, the collection and dispatch time is the time difference from the creation of the order to the start of processing by the salesperson. The collection and dispatch time of each order is obtained and compared with the preset time threshold. If the actual collection and dispatch time exceeds the preset threshold, it means that there is a delay in the collection and dispatch of the order. If the actual collection and dispatch time does not exceed the preset threshold, it means that there is no delay in the collection and dispatch of the order.

[0071] 215. If yes, then analyze the salesperson with high delay rate and the business dispatch area information to obtain the delay cause analysis result;

[0072] In this embodiment, if the delivery time of the unfinished order information exceeds the preset time threshold, the salesperson with high delay rate and the business delivery area information are analyzed to obtain the delay cause analysis results. The delay cause analysis results include excessive workload, processing efficiency, insufficient skills, insufficient support, geographical location, traffic conditions, customer communication efficiency, and insufficient salespersons in the area.

[0073] In the embodiment of the present invention, the accuracy and completeness of the final data are ensured through steps such as data cleaning, correction and filling of missing values, and data from different sources or formats are unified into a preset standard format to ensure consistency between different data sets. An advanced method with time series modeling capability is adopted to capture the historical behavior patterns of salesmen and collection and dispatch areas, and improve prediction accuracy. The multi-stage verification process can effectively avoid overfitting, improve the generalization ability of the model, and ensure its stability and accuracy in practical applications. It can dynamically identify salesmen and areas with high delay rates, deeply analyze the causes of high delays, and further identify the root causes of the problem by analyzing the information of unfinished orders and timed orders, helping salesmen and management to handle tasks more efficiently, improve operational efficiency, and make timely responses in terms of customer experience. By predicting and analyzing business collection and dispatch delays, it can reduce customer waiting time, improve service quality, and thus enhance customer satisfaction and loyalty.

[0074] See also Figure 3 The third embodiment of the intelligent expediting method for logistics orders in the embodiment of the present invention includes:

[0075] 301. Use genetic algorithm or ant colony optimization algorithm to build a path planning model;

[0076] In this embodiment, both the genetic algorithm (GA) and the ant colony optimization algorithm (ACO) are heuristic algorithms suitable for solving complex combinatorial optimization problems. The genetic algorithm or the ant colony optimization algorithm is used to construct a path planning model.

[0077] 302. Input the information of the salesperson with high delay rate and the delivery area and the delay cause analysis result into the path planning model to perform path planning, and obtain the specific delivery route, delivery sequence and estimated delivery time;

[0078] In this embodiment, the information of salesmen with high delay rates, business pick-up and dispatch areas, and the results of delay cause analysis are input into the path planning model. This process can use a genetic algorithm (GA) or an ant colony optimization algorithm (ACO) to weight the path selection according to the cause of the delay, optimize the final pick-up and dispatch path, sequence, and estimated time, and obtain a specific pick-up and dispatch route, pick-up and dispatch sequence, and estimated pick-up and dispatch time.

[0079] 303. Integrate the specific pick-up and dispatch driving route, pick-up and dispatch sequence, and estimated pick-up and dispatch time into a data structure to obtain pick-up and dispatch route planning information;

[0080] In this embodiment, a main data structure containing multiple substructures is designed. The main structure can be represented by a dictionary (or object) and organized according to the ID of each salesperson. Each salesperson's ID contains information such as the path, sequence, and estimated time of his task.

[0081] 304. Generate a visualized pick-up and dispatch path planning chart based on the pick-up and dispatch path planning information;

[0082] In this embodiment, relevant task data is extracted from the data structure, and the geographical location of the task is processed (the address is converted into longitude and latitude coordinates). According to the order of tasks and the coordinates of each task, a salesperson's delivery route map is generated. The salesperson's delivery route is drawn according to the order of the path, and the estimated completion time of each task is displayed on the timeline, or different priorities or time periods are distinguished by color. On the path map, each task point can use a different color to indicate the estimated time. For example, early tasks are represented by green and later tasks are represented by red. Thus, an interactive map is generated that can be opened in a browser to display the path and task points.

[0083] 305. Send the route planning chart for pickup and dispatch to the salesperson's terminal;

[0084] In this embodiment, a message push system is used to send the pick-up and dispatch route planning chart to the salesperson's terminal.

[0085] 306. Encrypt the pick-up and dispatch path planning graph to obtain the pick-up and dispatch path planning encrypted graph, and upload the pick-up and dispatch path planning encrypted graph to the blockchain;

[0086] In this embodiment, the dispatch path planning chart is encrypted by symmetric encryption (such as AES) or asymmetric encryption (such as RSA) to obtain an encrypted dispatch path planning chart, and a blockchain platform suitable for storing files or data is selected. The purpose of uploading the encrypted path chart to the blockchain is to ensure that the data cannot be tampered with and is traceable.

[0087] In the embodiment of the present invention, by using a genetic algorithm or an ant colony optimization algorithm to construct a path planning model, it is possible to efficiently search for the optimal or near-optimal solution in large-scale path selection, effectively avoid the shortcomings of manual planning, reduce labor costs, and provide more efficient and economical route selection. The information of salesmen with high delay rates and business pick-up and dispatch areas and the results of delay cause analysis are input into the path planning model, which can more accurately respond to complex situations in reality, improve pick-up and dispatch efficiency, reduce delays, and thus improve overall business operations. Through automated path planning and scheduling, salesmen no longer need to manually adjust routes or sequences. The system can generate the optimal route and pickup and dispatch sequence in real time based on the dynamically changing salesperson and regional information. This not only reduces the interference of human factors and improves work efficiency, but also ensures timely response to emergencies, such as salesperson delays and sudden traffic conditions. It generates a visual pickup and dispatch route planning chart, so that managers and salespeople can intuitively understand the pickup and dispatch routes and task arrangements. Encryption can ensure the confidentiality and tamper-proofness of route planning information, and prevent sensitive data from being maliciously accessed or modified. The use of blockchain technology further enhances the credibility and transparency of the data, ensuring that every step in the business process can be traced, verified and audited.

[0088] See also Figure 4 The fourth embodiment of the intelligent expediting method for logistics orders in the embodiment of the present invention includes:

[0089] 401. Create a listener and use the listener to monitor the completion status of the pickup and dispatch route planning information;

[0090] In this embodiment, an event-driven model is adopted, and a listener is created using a message queue or event bus. When the path planning information is completed, the system triggers an event through an API or database status change. The system pushes the event information to the listener through the message queue, passing the relevant task status, timestamp, and path information, etc.

[0091] 402. When the listener detects the completion status, it obtains the salesperson's dispatch completion information corresponding to the current completion status;

[0092] In this embodiment, when the listener listens to the completion status, it obtains the salesperson's dispatch completion information corresponding to the current completion status. Once the listener receives the task completion event, it needs to parse the message content and extract information related to the salesperson's dispatch work.

[0093] 403. Evaluate the path optimization effect of the salesperson's completion information and obtain an evaluation score;

[0094] In this embodiment, the key dimensions of evaluation are clearly defined, including task completion timeliness, path optimization effect, task success rate, customer feedback, and resource utilization efficiency. Based on the above dimensions, specific evaluation indicators can be designed, and the scores of each dimension can be weighted to obtain a comprehensive score. The calculated comprehensive score is used as the evaluation score of the salesperson's path optimization effect.

[0095] 404. Determine whether the evaluation score reaches a preset evaluation threshold;

[0096] In this embodiment, for each salesperson, after completing the task, a comprehensive score is calculated through the scores of the above-mentioned dimensions. Once the comprehensive score is obtained, it is necessary to determine whether the score reaches the preset evaluation threshold.

[0097] 405. If not, generate model adjustment parameters;

[0098] In this embodiment, if the evaluation score reaches a preset evaluation threshold, an automated hyperparameter tuning tool (such as the AutoML framework) is used to allow the system to automatically search for the best parameter combination and generate model adjustment parameters.

[0099] 406. Iteratively update the path planning model using the model adjustment parameters;

[0100] In this embodiment, appropriate parameters are selected for adjustment based on the performance analysis results, the path planning model is iteratively updated using the model adjustment parameters, and the validation data set is used to evaluate its effect.

[0101] 407. When the listener fails to monitor the completion status within the preset time, an early warning message is generated and sent to the management terminal;

[0102] In this embodiment, when the listener fails to listen to the completion status within the preset time, an early warning message is generated. The early warning message includes a task identifier, timeout information, task status, reason prompt, and priority. Once the early warning message is generated, the next step is to send the information to the management terminal. The early warning message can be immediately transmitted to the management terminal through a real-time message push mechanism (such as WebSocket, MQTT, HTTP request, etc.). In addition to directly pushing the early warning message, the system can also send it through email, SMS or system notification. The management terminal is usually a centralized control interface responsible for monitoring, adjusting and responding to all tasks. The goal of sending early warning information to the management terminal is to ensure that relevant personnel can receive prompts of task abnormalities in a timely manner and can take necessary actions quickly.

[0103] In the embodiment of the present invention, the completion status of the route planning for pickup and dispatch is monitored in real time to ensure that the system can dynamically respond to the progress of the task, and can determine whether the model needs to be optimized and adjusted based on the evaluation score. If the evaluation result does not meet expectations, the path planning model is updated by automatically adjusting the parameters. This self-adjustment mechanism based on the evaluation result enables the path planning system to continuously improve and optimize as the data and actual execution status change, thereby improving the accuracy and efficiency of the planning. By combining the path planning with the completion status of the pickup and dispatch of the salesperson for real-time evaluation, the system can accurately quantify and analyze the task execution of the salesperson and adjust the model based on the actual results. If the listener does not detect the completion status within the preset time, the system will actively generate an early warning message and send it to the management terminal. This early warning mechanism ensures that the management can take timely measures before problems occur, thereby avoiding operational risks or customer experience problems caused by delayed path planning or the failure of the salesperson to complete the task in time.

[0104] The above describes the intelligent urging method for logistics orders in the embodiment of the present invention. The following describes the intelligent urging device for logistics orders in the embodiment of the present invention. Figure 5 In one embodiment of the present invention, an intelligent expediting device for logistics orders includes:

[0105] Extraction processing module 501, used to extract relevant data of all salesmen and business dispatching areas from the business system database, perform data preprocessing on the relevant data, and obtain standardized relevant data;

[0106] The prediction and sorting module 502 is used to predict the standardized related data using a preset delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, and to sort the salespersons and business pickup and dispatch areas according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result;

[0107] The screening and analysis module 503 is used to screen the salesmen and the business dispatching areas according to the preset ratio according to the sorting results, obtain the salesmen and business dispatching area information with high delay rates, analyze the salesmen and business dispatching area information with high delay rates, and obtain the delay cause analysis results;

[0108] A planning module 504 is used to construct a path planning model using a genetic algorithm or an ant colony optimization algorithm, input the information of the salesperson with high delay rate and the business pick-up and dispatch area and the delay cause analysis result into the path planning model to perform path planning, and obtain pick-up and dispatch path planning information;

[0109] The uploading and sending module 505 is used to upload the pick-up and dispatch path planning information to the blockchain and send the pick-up and dispatch path planning information to the salesperson terminal.

[0110] In this embodiment, the standardized relevant data is predicted by the delay prediction model to obtain the predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, the salesperson and the business pickup and dispatch area are sorted according to the predicted value of the pickup and dispatch delay rate, and the salesperson and the business pickup and dispatch area are screened according to the sorting result according to a preset ratio to obtain the information of the salesperson and the business pickup and dispatch area with high delay rate, and the information of the salesperson and the business pickup and dispatch area with high delay rate is analyzed to obtain the analysis result of the cause of delay, and the information of the salesperson and the business pickup and dispatch area with high delay rate and the analysis result of the cause of delay are input into the path planning model for path planning to obtain the pickup and dispatch path planning information, which can optimize the pickup and dispatch path for the salesperson and the business pickup and dispatch area with high delay rate, prevent the pickup and dispatch delay, and improve user satisfaction.

[0111] See also Figure 6 Another embodiment of the intelligent expediting device for logistics orders in the embodiment of the present invention includes:

[0112] Extraction processing module 501, used to extract relevant data of all salesmen and business dispatching areas from the business system database, perform data preprocessing on the relevant data, and obtain standardized relevant data;

[0113] The prediction and sorting module 502 is used to predict the standardized related data using a preset delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, and to sort the salespersons and business pickup and dispatch areas according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result;

[0114] The screening and analysis module 503 is used to screen the salesmen and the business dispatching areas according to the preset ratio according to the sorting results, obtain the salesmen and business dispatching area information with high delay rates, analyze the salesmen and business dispatching area information with high delay rates, and obtain the delay cause analysis results;

[0115] A planning module 504 is used to construct a path planning model using a genetic algorithm or an ant colony optimization algorithm, input the information of the salesperson with high delay rate and the business pick-up and dispatch area and the delay cause analysis result into the path planning model to perform path planning, and obtain pick-up and dispatch path planning information;

[0116] The uploading and sending module 505 is used to upload the pick-up and dispatch path planning information to the blockchain and send the pick-up and dispatch path planning information to the salesperson terminal;

[0117] In this embodiment, the extraction processing module 501 includes: an extraction unit 5011, which is used to extract relevant data of all salesmen and business dispatch areas from the business system database, and the relevant data includes salesman information, business dispatch area information and order information; a cleaning unit 5012, which is used to clean the salesman information, business dispatch area information and order information to obtain cleaned relevant data; a correction and filling unit 5013, which is used to correct abnormal values ​​of the cleaned relevant data to obtain corrected relevant data, and fill missing values ​​of the corrected relevant data to obtain filled relevant data; a conversion unit 5014, which is used to convert the format of the filled relevant data based on a preset target standard format to obtain standardized relevant data.

[0118] In this embodiment, the prediction and sorting module 502 includes: an acquisition unit 5021, which is used to acquire historical salesperson information, historical business pick-up and dispatch area information and historical order information; a first integration unit 5022, which is used to integrate the historical salesperson information, historical business pick-up and dispatch area information and historical order information to obtain historical related data; a division unit 5023, which is used to divide the historical related data according to a preset ratio to obtain a training set, a validation set and a test set; a selection and training unit 5024, which is used to select a recurrent neural network model, train the recurrent neural network model with the training set, and obtain a delay prediction model; a tuning and evaluation unit 5025, which is used to tune and evaluate the delay prediction model with the validation set and the test set respectively; a prediction unit 5026, which is used to use the delay prediction model to predict the standardized related data to obtain the pick-up and dispatch delay rate prediction value of each salesperson and business pick-up and dispatch area; a sorting unit 5027, which is used to sort the salesperson and the business pick-up and dispatch area according to the pick-up and dispatch delay rate prediction value to obtain a sorting result.

[0119] In this embodiment, the screening and analysis module 503 includes: a screening unit 5031, which is used to screen the salesmen and business dispatch areas according to a preset ratio based on the sorting results, and obtain the information of the salesmen and business dispatch areas with high delay rates; an identification unit 5032, which is used to identify the unfinished order information in the information of the salesmen and business dispatch areas with high delay rates; a judgment unit 5033, which is used to judge whether the dispatch time of the unfinished order information exceeds a preset time threshold; and an analysis unit 5034, which is used to analyze the information of the salesmen and business dispatch areas with high delay rates if so, and obtain the delay cause analysis results.

[0120] In this embodiment, the construction planning module 504 includes: a construction unit 5041, which is used to construct a path planning model using a genetic algorithm or an ant colony optimization algorithm; a planning unit 5042, which is used to input the information of salesmen with high delay rates and business pick-up and dispatch areas and the results of delay cause analysis into the path planning model for path planning, and obtain a specific pick-up and dispatch driving path, a pick-up and dispatch sequence, and an estimated pick-up and dispatch time; a second integration unit 5043, which is used to integrate the specific pick-up and dispatch driving path, a pick-up and dispatch sequence, and an estimated pick-up and dispatch time into a data structure to obtain the pick-up and dispatch path planning information.

[0121] In this embodiment, the upload and sending module 505 includes: a generation unit 5051, which is used to generate a visual pick-up and dispatch path planning chart based on the pick-up and dispatch path planning information; a sending unit 5052, which is used to send the pick-up and dispatch path planning chart to the salesperson terminal; an encryption upload unit 5053, which is used to encrypt the pick-up and dispatch path planning chart to obtain an encrypted pick-up and dispatch path planning chart, and upload the encrypted pick-up and dispatch path planning chart to the blockchain.

[0122] In this embodiment, it also includes: creating a monitoring module 506, which is used to create a listener and use the listener to monitor the completion status of the pick-up and dispatch path planning information; an acquisition module 507, which is used to obtain the salesperson's pick-up and dispatch completion information corresponding to the current completion status when the listener monitors the completion status; an evaluation module 508, which is used to evaluate the path optimization effect of the salesperson's pick-up and dispatch completion information to obtain an evaluation score; a judgment module 509, which is used to judge whether the evaluation score reaches a preset evaluation threshold; a generation module 510, which is used to generate model adjustment parameters if not; an update module 511, which is used to iteratively update the path planning model using the model adjustment parameters; and a generation and sending module 512, which is used to generate warning information when the listener fails to monitor the completion status within a preset time, and send the warning information to the management terminal.

[0123] above Figure 5 and Figure 6 The intelligent dispatch device for logistics orders in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The intelligent dispatch device for logistics orders in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0124] Figure 7: is a structural schematic diagram of an intelligent expediting device for logistics orders provided in an embodiment of the present invention. The intelligent expediting device 600 for logistics orders may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 610 (for example, one or more processors) and a memory 620, and one or more storage media 630 (for example, one or more mass storage devices) storing application programs 633 or data 632. Among them, the memory 620 and the storage medium 630 can be short-term storage or permanent storage. The program stored in the storage medium 630 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the intelligent expediting device 600 for logistics orders. Furthermore, the processor 610 can be configured to communicate with the storage medium 630, and execute a series of instruction operations in the storage medium 630 on the intelligent expediting device 600 for logistics orders to implement the steps of the intelligent expediting method for logistics orders provided in the above-mentioned method embodiments.

[0125] The intelligent expediting device 600 for logistics orders may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input and output interfaces 660, and / or one or more operating systems 631, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Figure 7 The structure of the intelligent dispatching equipment for logistics orders shown does not constitute a limitation on the intelligent dispatching equipment based on logistics orders, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0126] 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 an intelligent dispatching method for logistics orders.

[0127] 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.

[0128] 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 all 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 a number of instructions to enable a computer device (which can be a personal computer, a server, or a 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 codes.

[0129] Finally, it should be noted that the above description is only a preferred example of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent expediting method for logistics orders, characterized in that: include: Extract relevant data of all salesmen and business dispatch areas from the business system database, perform data preprocessing on the relevant data, and obtain standardized relevant data; The standardized related data are predicted using a preset delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each of the salespersons and the business pickup and dispatch area, and the salespersons and the business pickup and dispatch areas are sorted according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result; According to the sorting results, the salespersons and the business dispatching areas are screened according to a preset ratio to obtain information about the salespersons and business dispatching areas with high delay rates, and the salespersons and business dispatching areas with high delay rates are analyzed to obtain delay cause analysis results; A path planning model is constructed by using a genetic algorithm or an ant colony optimization algorithm, and the information of the salesperson with a high delay rate and the business pick-up and dispatch area and the delay cause analysis result are input into the path planning model for path planning to obtain pick-up and dispatch path planning information; The route planning information for pickup and dispatch is uploaded to the blockchain, and the route planning information for pickup and dispatch is sent to the salesperson's terminal.

2. The intelligent expediting method for logistics orders according to claim 1 is characterized in that: The relevant data of all salesmen and business dispatching areas are extracted from the business system database, and the relevant data are preprocessed to obtain standardized relevant data, including: Extract relevant data of all salesmen and business delivery areas from the business system database, wherein the relevant data includes salesman information, business delivery area information and order information; Performing data cleaning on the salesperson information, the business delivery area information, and the order information to obtain cleaned relevant data; Correcting abnormal values ​​of the cleaned relevant data to obtain corrected relevant data, and filling missing values ​​of the corrected relevant data to obtain filled relevant data; Based on a preset target standard format, the padded relevant data is format converted to obtain standardized relevant data.

3. The intelligent expediting method for logistics orders according to claim 1 is characterized in that: The standardized related data is predicted by using a preset delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each salesperson and business pickup and dispatch area, and the salespersons and business pickup and dispatch areas are sorted according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result, including: Obtain historical salesperson information, historical business delivery area information, and historical order information; Integrate the historical salesperson information, historical business delivery area information and historical order information to obtain historical related data; Dividing the historical related data according to a preset ratio to obtain a training set, a validation set, and a test set; Selecting a recurrent neural network model, and training the recurrent neural network model using the training set to obtain a delay prediction model; Tuning and evaluating the delay prediction model respectively through the validation set and the test set; The standardized related data is predicted using the delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each of the salespersons and the business pickup and dispatch area; The salespersons and business delivery areas are sorted according to the delivery delay rate prediction value to obtain a sorting result.

4. The intelligent expediting method for logistics orders according to claim 1 is characterized in that: The salespersons and the business dispatching areas are screened according to the sorting results according to a preset ratio to obtain information about salespersons and business dispatching areas with high delay rates, and the salespersons and business dispatching areas with high delay rates are analyzed to obtain delay cause analysis results, including: According to the sorting results, the salespersons and the business dispatching areas are screened according to a preset ratio to obtain information on the salespersons and business dispatching areas with high delay rates; Identify the unfinished order information in the salesperson with high delay rate and the business delivery area information; Determine whether the delivery time of the uncompleted order information exceeds a preset time threshold; If so, the salesperson with high delay rate and the business dispatch area information are analyzed to obtain the delay cause analysis result.

5. The intelligent expediting method for logistics orders according to claim 1 is characterized in that: The genetic algorithm or ant colony optimization algorithm is used to construct a path planning model, and the information of the salesperson with high delay rate and the business pick-up and dispatch area and the delay cause analysis result are input into the path planning model for path planning to obtain the pick-up and dispatch path planning information, including: Use genetic algorithm or ant colony optimization algorithm to build path planning model; Input the salesperson with high delay rate and the delivery area information and the delay cause analysis result into the path planning model to perform path planning, and obtain the specific delivery route, delivery sequence and estimated delivery time; The specific pickup and dispatch driving route, the pickup and dispatch order, and the estimated pickup and dispatch time are integrated into a data structure to obtain the pickup and dispatch route planning information.

6. The intelligent expediting method for logistics orders according to claim 1 is characterized in that: The uploading of the pick-up and dispatch path planning information to the blockchain and sending the pick-up and dispatch path planning information to the salesperson terminal includes: Generate a visual pick-up and dispatch path planning chart according to the pick-up and dispatch path planning information; Sending the route planning chart to the salesperson's terminal; The pick-up and dispatch path planning chart is encrypted to obtain a pick-up and dispatch path planning encrypted chart, and the pick-up and dispatch path planning encrypted chart is uploaded to the blockchain.

7. The intelligent expediting method for logistics orders according to claim 1 is characterized in that: After uploading the route planning information to the blockchain and sending the route planning information to the salesperson terminal, the method further includes: Create a listener, and use the listener to monitor the completion status of the pickup and dispatch route planning information; When the listener monitors the completion status, the salesperson's dispatch completion information corresponding to the current completion status is obtained; Evaluate the path optimization effect of the salesperson's completion information and obtain an evaluation score; Determining whether the evaluation score reaches a preset evaluation threshold; If not, then generate model adjustment parameters; Iteratively updating the path planning model using the model adjustment parameters; When the listener fails to monitor the completion status within a preset time, an early warning message is generated and sent to the management terminal.

8. An intelligent expediting device for logistics orders, characterized in that: include: An extraction processing module is used to extract relevant data of all salesmen and business dispatch areas from the business system database, perform data preprocessing on the relevant data, and obtain standardized relevant data; A prediction and sorting module is used to predict the standardized related data using a preset delay prediction model to obtain a predicted value of the pickup and dispatch delay rate for each of the salespersons and the business pickup and dispatch area, and to sort the salespersons and the business pickup and dispatch areas according to the predicted value of the pickup and dispatch delay rate to obtain a sorting result; A screening and analysis module, for screening the salespersons and the business dispatching areas according to a preset ratio according to the sorting results, obtaining information on the salespersons and business dispatching areas with high delay rates, analyzing the information on the salespersons and business dispatching areas with high delay rates, and obtaining delay cause analysis results; Constructing a planning module, which is used to construct a path planning model using a genetic algorithm or an ant colony optimization algorithm, inputting the information of the salesperson with high delay rate and the business pick-up and dispatch area and the delay cause analysis result into the path planning model for path planning, and obtaining pick-up and dispatch path planning information; The uploading and sending module is used to upload the pick-up and dispatch path planning information to the blockchain, and send the pick-up and dispatch path planning information to the salesperson terminal.

9. An intelligent expediting device for logistics orders, characterized in that: The intelligent expediting device for logistics orders includes: a memory and at least one processor, wherein instructions are stored in the memory; At least one of the processors calls the instructions in the memory so that the intelligent dispatching device for logistics orders executes each step of the intelligent dispatching method for logistics orders 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 intelligent expediting method for logistics orders as described in any one of claims 1-7 are implemented.

Images