Intelligent distribution method, device and equipment for logistics orders

Abstract

The invention discloses an intelligent distribution method, device and equipment for logistics orders, and the method is applied to the technical field of logistics, and the method comprises the steps: obtaining historical logistics order data, and determining a historical distribution result of the historical logistics order data; inputting the historical logistics order data into a preset model, and performing order distribution processing on the historical logistics order data to obtain a plurality of order distribution results; obtaining order distribution weights corresponding to the address salesman order distribution algorithm, the parcel collection code electronic fence algorithm and the website fixed salesman algorithm; determining a predicted order distribution result according to the plurality of order distribution results and the plurality of weights; obtaining an intelligent order distribution model according to the difference between the predicted order distribution result and the historical order distribution result; and inputting the to-be-processed logistics order into the intelligent order distribution model for order distribution processing to obtain a target order distribution algorithm and an order distribution result. According to the invention, the source of the order distribution decision is quickly traced, and intelligent, accurate and quick order distribution of the logistics order is realized.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device and equipment for intelligently splitting logistics orders. Background Art

[0002] In the logistics industry, the accuracy of order splitting is crucial to improving logistics efficiency, reducing costs and increasing customer satisfaction. Traditional automatic order splitting systems have many problems, such as low order splitting accuracy, difficulty in distinguishing the causes of mis-splitting, opaque order splitting methods, order splitting algorithms that cannot fully exert their advantages, and lack of effective data tracing and analysis methods. These problems lead to adverse consequences such as order allocation errors, low processing efficiency, and increased customer complaints during logistics operations.

[0003] Due to the complex logic of order splitting, users cannot understand why a certain result is calculated in the order splitting. In order to make the calculation result traceable and facilitate users to intervene in the calculation result, it is necessary to display the key calculation basis to the user so that the calculation basis can be intervened or the cause of the problem can be confirmed. Summary of the invention

[0004] The present invention provides a method, device and equipment for intelligent order splitting of logistics orders, which can quickly trace the source of order splitting decisions and realize intelligent, accurate and rapid order splitting of logistics orders.

[0005] In one aspect, the present invention provides a method for intelligently splitting logistics orders, the method comprising: Obtaining historical logistics order data, and determining historical order splitting results of the historical logistics order data; The historical logistics order data is input into a preset model, and the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the network point fixed salesperson algorithm are respectively used to split the historical logistics order data to obtain a first order splitting result corresponding to the address salesperson order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the network point fixed salesperson algorithm; Obtain a first order distribution weight corresponding to the address salesperson order distribution algorithm, a second order distribution weight corresponding to the pickup code electronic fence algorithm, and a third order distribution weight corresponding to the network point fixed salesperson algorithm; Determine a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; According to the difference between the predicted order splitting result and the historical order splitting result, the preset model is trained to obtain an intelligent order splitting model; Obtain the pending logistics orders, input the pending logistics orders into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting results; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm.

[0006] In an exemplary embodiment, the obtaining of historical logistics order data includes: Get initial logistics order data; Using a data cleaning algorithm to clean the initial logistics order data, remove noise and erroneous information, and obtain cleaned order data; A standardized algorithm is used to convert the address, customer information, and product information in the cleaned order data into a unified format to obtain the historical logistics order data.

[0007] In an exemplary embodiment, the method further comprises: Adopting the address splitting strategy to extract the historical order number and historical address information of the historical logistics order data; Obtaining the historical pickup clerk corresponding to the historical order number, and constructing a corresponding relationship between the historical address information and the historical pickup clerk to form a first database; The historical keywords, historical outlets, and historical pickup codes of the historical logistics order data are extracted using a pickup code keyword strategy, and a corresponding relationship between the historical keywords and the historical outlets and the historical pickup codes is constructed to form a second database; The pickup code electronic fence strategy is adopted to extract the historical longitude and latitude, historical outlets and historical pickup codes of the historical logistics order data, and determine the historical electronic fence corresponding to the historical longitude and latitude; A correspondence between the historical electronic fences, the historical outlets, and the historical pickup codes is constructed to form a third database.

[0008] In an exemplary embodiment, the method further comprises: Acquire the real-time address information of the order data to be processed, search the first database for a pickup agent corresponding to the real-time address information, and obtain a target pickup agent; or Obtaining a real-time keyword of the order data to be processed, and searching the second database for a real-time outlet and a real-time pickup code corresponding to the real-time keyword; or Obtaining the real-time longitude and latitude of the pending order data, and determining the real-time electronic fence corresponding to the real-time longitude and latitude; The real-time network point and the real-time pickup code corresponding to the real-time electronic fence are searched in the third database.

[0009] In an exemplary embodiment, determining a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight includes: Calculate a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; The sum of the first product, the second product and the third product is calculated to obtain the predicted order splitting result.

[0010] In an exemplary embodiment, the method further comprises: Determine the real-time priority of each logistics outlet based on the real-time order backlog of each logistics outlet in the logistics platform; Determine the real-time priority of each salesperson according to the real-time working status of each salesperson in the logistics platform; Determine the real-time priority of each pending order data according to the real-time order processing status of each pending order data in the physical platform; According to the real-time priority of each logistics network point, the real-time priority of each salesperson and the real-time priority of each pending order data, the target network point and target salesperson for each pending order data are determined.

[0011] In an exemplary embodiment, the method further comprises: Obtaining logistics order data whose historical order splitting results are erroneous results from the plurality of historical logistics order data to obtain wrongly split logistics order data; Input the misclassified logistics order data and the order splitting results into a machine learning model to extract order features, order splitting algorithm execution features, and external environment features; Identify the results of misclassification factors based on the order characteristics, order splitting algorithm execution characteristics, and external environment characteristics; According to the difference between the misclassification factor result and the real misclassification factor of the misclassified logistics order data, the machine learning model is trained to obtain a misclassification cause analysis model; Inputting the pending order data and the real-time order splitting results into the misclassification cause analysis model to perform misclassification factor analysis and obtain target misclassification factors; According to the target misclassification factor, an early warning prompt message is generated or the real-time order splitting result is adjusted according to the target misclassification factor to generate a revised order splitting result.

[0012] On the other hand, a smart order splitting device for logistics orders is provided, the device comprising: A historical order splitting result determination module is used to obtain historical logistics order data and determine the historical order splitting results of the historical logistics order data; An order splitting result determination module is used to input the historical logistics order data into a preset model, and respectively adopt an address salesperson order splitting algorithm, a pickup code electronic fence algorithm, and a network fixed salesperson algorithm to split the historical logistics order data, and obtain a first order splitting result corresponding to the address salesperson order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the network fixed salesperson algorithm; An order splitting weight determination module is used to obtain a first order splitting weight corresponding to the address salesperson order splitting algorithm, a second order splitting weight corresponding to the pickup code electronic fence algorithm, and a third order splitting weight corresponding to the fixed salesperson algorithm at the outlet; An order splitting result prediction module, used to determine a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; An intelligent order splitting model training module, used to train the preset model according to the difference between the predicted order splitting result and the historical order splitting result to obtain an intelligent order splitting model; The order splitting processing module is used to obtain the logistics orders to be processed, input the logistics orders to be processed into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting results; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm.

[0013] In an exemplary embodiment, the historical order splitting result determination module includes: A data acquisition unit, used to acquire initial logistics order data; A cleaning unit, used to clean the initial logistics order data using a data cleaning algorithm to remove noise and error information to obtain cleaned order data; The data unification unit is used to convert the address, customer information, and product information in the cleaned order data into a unified format using a standardized algorithm to obtain the historical logistics order data.

[0014] In an exemplary embodiment, the apparatus further comprises: A historical information extraction module, used to extract the historical order number and historical address information of the historical logistics order data by adopting an address splitting strategy; A first database construction module is used to obtain the historical pickup salesperson corresponding to the historical order number, and to construct a corresponding relationship between the historical address information and the historical pickup salesperson to form a first database; A second database construction module is used to extract historical keywords, historical outlets and historical pickup codes of the historical logistics order data using a pickup code keyword strategy, and to construct a corresponding relationship between the historical keywords and the historical outlets and the historical pickup codes to form a second database; An electronic fence determination module, used to extract the historical longitude and latitude, historical outlets and historical pickup codes of the historical logistics order data by adopting the pickup code electronic fence strategy, and determine the historical electronic fence corresponding to the historical longitude and latitude; The third database construction module is used to construct the correspondence between the historical electronic fences and the historical outlets and the historical collection codes to form a third database.

[0015] In an exemplary embodiment, the apparatus further comprises: A first information acquisition module is used to acquire the real-time address information of the order data to be processed, and to search the first database for a pickup agent corresponding to the real-time address information to obtain a target pickup agent; or A second information acquisition module is used to acquire the real-time keywords of the order data to be processed, and search the second database for the real-time outlets and real-time pickup codes corresponding to the real-time keywords; or A third information acquisition module is used to obtain the real-time longitude and latitude of the pending order data and determine the real-time electronic fence corresponding to the real-time longitude and latitude; The data search module is used to search the real-time network point and the real-time pickup code corresponding to the real-time electronic fence in the third database.

[0016] In an exemplary embodiment, the order splitting result prediction module includes: A first calculation unit, used for calculating a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; The second calculation unit is used to calculate the sum of the first product, the second product and the third product to obtain the predicted order splitting result.

[0017] In an exemplary embodiment, the apparatus further comprises: A first priority determination module is used to determine the real-time priority of each logistics outlet according to the real-time order backlog of each logistics outlet in the logistics platform; A second priority determination module is used to determine the real-time priority of each salesperson according to the real-time working status of each salesperson in the logistics platform; A third priority determination module, configured to determine the real-time priority of each to-be-processed order data according to the real-time order processing status of each to-be-processed order data in the physical platform; The data determination module is used to determine the target outlet and target salesperson to which each pending order data is assigned based on the real-time priority of each logistics outlet, the real-time priority of each salesperson, and the real-time priority of each pending order data.

[0018] In an exemplary embodiment, the apparatus further comprises: A wrongly sorted data acquisition module is used to acquire logistics order data whose historical order splitting results are wrong results from the plurality of historical logistics order data to obtain the wrongly sorted logistics order data; A feature extraction module is used to input the misclassified logistics order data and the order splitting results into a machine learning model to extract order features, order splitting algorithm execution features and external environment features; A result identification module, used to identify the result of misclassification factors according to the order characteristics, the execution characteristics of the order splitting algorithm and the external environment characteristics; A model training module, used to train the machine learning model according to the difference between the misclassification factor result and the real misclassification factor of the misclassified logistics order data, so as to obtain a misclassification cause analysis model; A factor determination module, used for inputting the to-be-processed order data and the real-time order splitting results into the misclassification cause analysis model to perform misclassification factor analysis and obtain target misclassification factors; The correction module is used to generate early warning prompt information according to the target misclassification factor or adjust the real-time order splitting result according to the target misclassification factor to generate a corrected order splitting result.

[0019] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the intelligent order splitting method for logistics orders as described above.

[0020] On the other hand, a computer storage medium is provided, which stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the intelligent order splitting method for logistics orders as described above.

[0021] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes to implement the intelligent order splitting method for logistics orders as described above.

[0022] The intelligent order splitting method, device and equipment for logistics orders provided by the present invention have the following technical effects: The present invention obtains historical logistics order data, and determines historical order splitting results of the historical logistics order data; inputs the historical logistics order data into a preset model, and respectively adopts an address clerk order splitting algorithm, a pickup code electronic fence algorithm, and a network fixed clerk algorithm to split the historical logistics order data, and obtains a first order splitting result corresponding to the address clerk order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the network fixed clerk algorithm; obtains a first order splitting weight corresponding to the address clerk order splitting algorithm, a second order splitting weight corresponding to the pickup code electronic fence algorithm, and a third order splitting weight corresponding to the network fixed clerk algorithm. Order splitting weight; determine the predicted order splitting result according to the first product of the first order splitting result and the first order splitting weight, the second product of the second order splitting result and the second order splitting weight, and the third product of the third order splitting result and the third order splitting weight; train the preset model according to the difference between the predicted order splitting result and the historical order splitting result to obtain an intelligent order splitting model; obtain the pending logistics order, input the pending logistics order into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting result; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the fixed salesperson algorithm of the outlet. The present invention can quickly trace back to the source of the order splitting decision, can quickly trace back to the source of the order splitting decision, and realize the intelligent, accurate and fast order splitting of logistics orders. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions and advantages of the embodiments of this specification or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 It is a schematic diagram of an intelligent order distribution system for logistics orders provided in an embodiment of this specification; Figure 2It is a flowchart of an intelligent order splitting method for logistics orders provided in an embodiment of this specification; Figure 3 It is a flowchart of a method for obtaining historical logistics order data provided by an embodiment of this specification; Figure 4 It is a flowchart of a method for searching a real-time outlet and a real-time pickup code provided in an embodiment of this specification; Figure 5 It is a flowchart of a method for determining a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight, provided in an embodiment of this specification; Figure 6 It is a structural schematic diagram of an intelligent order splitting device for logistics orders provided in an embodiment of this specification; Figure 7 It is a structural diagram of a server provided in an embodiment of this specification. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. 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 interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0027] See also Figure 1 , Figure 1 is a schematic diagram of an intelligent order distribution system for logistics orders provided in an embodiment of this specification, such as Figure 1 As shown, the intelligent order splitting system for the logistics order may include at least a server 01 and a client 02.

[0028] Specifically, in the embodiments of this specification, the server 01 may include an independently operated server, or a distributed server, or a server cluster composed of multiple servers, and may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Server 01 may include a network communication unit, a processor, a memory, and the like. Specifically, the server 01 can be used for training to obtain an intelligent order distribution model.

[0029] Specifically, in the embodiments of this specification, the client 02 may include physical devices such as smart phones, desktop computers, tablet computers, laptop computers, digital assistants, smart wearable devices, smart speakers, vehicle terminals, smart TVs, etc., and may also include software running in physical devices, such as web pages provided by some service providers to users, or applications provided by these service providers to users. Specifically, the client 02 may be used to query the target order splitting algorithm and order splitting results of the logistics orders to be processed according to the intelligent order splitting model.

[0030] The following is an intelligent order splitting method for logistics orders of the present invention. Figure 2 It is a flowchart of a method for intelligent order splitting of logistics orders provided in an embodiment of this specification. This specification provides method operation steps as described in the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The sequence of steps listed in the embodiment is only one way of executing the steps among many sequences, and does not represent the only execution sequence. When the actual system or server product is executed, it can be executed in sequence or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method shown in the embodiment or the drawings. Specifically, Figure 2 As shown, the method may include: S201: Acquire historical logistics order data, and determine historical order splitting results of the historical logistics order data. In the embodiments of the present specification, the historical logistics order data may be data of historical logistics orders in the logistics platform, and the historical order splitting result may be a result obtained by splitting the historical logistics order data according to the order splitting strategy.

[0031] S203: Input the historical logistics order data into a preset model, and use the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm to split the historical logistics order data, and obtain a first order splitting result corresponding to the address salesperson order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the branch fixed salesperson algorithm.

[0032] In the embodiments of this specification, the order splitting process can be performed separately according to multiple algorithms in the preset model to obtain the order splitting results corresponding to each algorithm; multiple algorithms can be used to integrate the order splitting: an intelligent order splitting model that integrates multiple order splitting algorithms is developed. The model comprehensively considers multiple factors such as the address clerk sedimentation data, the pickup code electronic fence algorithm, and the branch fixed clerk algorithm to make order splitting decisions. For example, the decision tree model in the machine learning algorithm is used to train the optimal weights of different order splitting algorithms in different scenarios based on historical order data, so as to automatically select the most appropriate order splitting algorithm or algorithm combination according to the order characteristics.

[0033] S205: Obtain a first order distribution weight corresponding to the address salesperson order distribution algorithm, a second order distribution weight corresponding to the pickup code electronic fence algorithm, and a third order distribution weight corresponding to the network point fixed salesperson algorithm.

[0034] In the embodiments of the present specification, the weights of multiple algorithms can be predicted according to the preset model, and the order distribution weights corresponding to each algorithm can be obtained in turn.

[0035] S207: Determine a predicted order splitting result based on a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight.

[0036] In the embodiments of the present specification, the predicted order splitting result can be obtained by integrating the order splitting results of various algorithms.

[0037] S209: According to the difference between the predicted order splitting result and the historical order splitting result, the preset model is trained to obtain an intelligent order splitting model.

[0038] In the embodiment of the present specification, after determining the predicted order splitting result, the target loss data can be determined according to the difference between the predicted order splitting result and the historical order splitting result, and the model parameters of the preset model can be adjusted according to the target loss data until the training end condition is met, and the preset model at the end of the training is determined as the intelligent order splitting model. The training end condition can be determined according to the target loss data or the number of training iterations.

[0039] S2011: Obtain the logistics order to be processed, input the logistics order to be processed into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting results; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm.

[0040] In an embodiment of the present specification, after the intelligent order splitting model is trained, the logistics orders to be processed can be input into the model, and at least one target order splitting algorithm can be determined by predicting the weights of each algorithm; the target order splitting algorithm can be one algorithm or a combination of multiple algorithms, and then the order splitting is performed according to the target order splitting algorithm to obtain the corresponding order splitting results, thereby realizing intelligent and accurate order splitting.

[0041] In the embodiments of this specification, Figure 3 As shown, the acquisition of historical logistics order data includes: S301: Obtaining initial logistics order data; S303: Using a data cleaning algorithm to clean the initial logistics order data, remove noise and error information, and obtain cleaned order data; S305: Using a standardized algorithm, the address, customer information, and product information in the cleaned order data are converted into a unified format to obtain the historical logistics order data.

[0042] In the embodiment of this specification, order data preprocessing may include: Data cleaning: Design data cleaning algorithms to remove noise and erroneous information from order data. For example, establish a complex rule model for the logic of false order judgment. For orders that have been picked up but no order number has been returned and no customer or salesperson has reported cancellation, use a multi-dimensional data analysis model to determine whether it is a false order; for orders that have been picked up and the order number has been returned normally, further combine the logistics track analysis model of the waybill number to determine that the waybill number has no logistics track (and no customer or salesperson has reported cancellation records) or the waybill number has a logistics track but the date is earlier than the order date. Orders are false orders.

[0043] Data standardization: Use a standardized algorithm to convert the address, customer information, product information, etc. in the order data into a unified format to facilitate accurate processing by the subsequent order splitting algorithm. For example, use the address parsing model in natural language processing technology to parse address information into structured data, including detailed information such as province, city, district, and street, to improve the accuracy of address recognition.

[0044] In the embodiment of this specification, the method further includes: Adopting the address splitting strategy to extract the historical order number and historical address information of the historical logistics order data; Obtaining the historical pickup clerk corresponding to the historical order number, and constructing a corresponding relationship between the historical address information and the historical pickup clerk to form a first database; The historical keywords, historical outlets, and historical pickup codes of the historical logistics order data are extracted using a pickup code keyword strategy, and a corresponding relationship between the historical keywords and the historical outlets and the historical pickup codes is constructed to form a second database; The pickup code electronic fence strategy is adopted to extract the historical longitude and latitude, historical outlets and historical pickup codes of the historical logistics order data, and determine the historical electronic fence corresponding to the historical longitude and latitude; A correspondence between the historical electronic fences, the historical outlets, and the historical pickup codes is constructed to form a third database.

[0045] The calculation basis is defined as follows: i. Through different order splitting methods such as historical learning, longitude and latitude, keywords, etc., when the system outputs a certain result, it can trace back to the core data of the previous node and calculate the final result according to the preset rules.

[0046] ii. For example: 1. Address order splitting: learn from a certain order number that a certain order at this address has been handled by a certain salesperson If the order is received, the same address will be assigned to the last salesperson next time; therefore, the calculation basis is a certain order number; 2. Pickup code keywords: a province, city or district maintained by the business + keywords to predict the outlets + pickup Code; when the corresponding keyword can be parsed from the user's order address, the corresponding outlet and pickup code can be output, and the calculation basis is: keyword; 3. Pickup code electronic fence: map interface. The map interface returns the result, including the order splitting method, order splitting result, and calculation basis [the order address is parsed into longitude and latitude, and then the longitude and latitude are matched with the electronic fence to output the outlet + pickup code] The present invention can display the calculation basis, so that users can find the root cause of the problem in time and deal with the problem in a targeted manner; a. Output fields of the optimal prediction interface for pickup code: i. Order number: Take the waybill number / partner order number / order number; according to the existing rules, take the global common value; to facilitate problem tracing and location.

[0047] ii. Predicted outlets: Automatically sort orders and output outlet codes iii. Predicted pickup code: Automatically output the pickup code for each order. If there is no pickup code, this field will be empty. iv. Predicted salesperson: the salesperson found by searching the outlet + pickup code or the salesperson directly returned by the interface; v. Terminal code: map interface return value; vi. Grouping: map interface return value; vii. Package stack: map interface return value; viii. Order splitting algorithm: current order splitting algorithm; ix. Calculation basis: x. Package pile prediction type: map interface return value xi. Package pile prediction basis: map interface return value b. Calculation based on rule output logic: i. Historical learning algorithms: address sorting, intelligent models, etc.; output, calculation basis is unified as: order number, that is, the learned order number; ii. Rule type: Pickup code keyword algorithm, the calculation basis is: keyword; iii. Map algorithm output: 1. type=mapgeo, the calculation basis is unified as: longitude and latitude; 2. If type=package pile, the package pile prediction type and package pile prediction basis (provided by the map interface) are additionally output.

[0048] In the embodiments of this specification, Figure 4 As shown, the method also includes: S401: Acquire real-time address information of the order data to be processed, search the first database for a pickup agent corresponding to the real-time address information, and obtain a target pickup agent; or S403: Acquire the real-time keyword of the order data to be processed, and search the second database for the real-time outlet and the real-time pickup code corresponding to the real-time keyword; or S405: Acquire the real-time longitude and latitude of the order data to be processed, and determine the real-time electronic fence corresponding to the real-time longitude and latitude; S407: Searching the third database for the real-time network point and the real-time pickup code corresponding to the real-time electronic fence.

[0049] In the embodiments of this specification, the optimal weights of different order splitting algorithms in different scenarios can be trained based on historical order data, so as to automatically select the most suitable order splitting algorithm or algorithm combination according to order characteristics.

[0050] In the embodiments of this specification, Figure 5 As shown, determining the predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight includes: S2071: Calculate a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; S2073: Calculate the sum of the first product, the second product and the third product to obtain the predicted order splitting result.

[0051] In the embodiment of this specification, the model comprehensively considers multiple factors such as the address salesperson sedimentation data, the pickup code electronic fence algorithm, the network point fixed salesperson algorithm, etc. to make order splitting decisions. Calculate the first product of the first order splitting result and the first order splitting weight, the second product of the second order splitting result and the second order splitting weight, and the third product of the third order splitting result and the third order splitting weight; then calculate the sum of the first product, the second product, and the third product to obtain the predicted order splitting result.

[0052] In the embodiment of this specification, the method further includes: Determine the real-time priority of each logistics outlet based on the real-time order backlog of each logistics outlet in the logistics platform; Determine the real-time priority of each salesperson according to the real-time working status of each salesperson in the logistics platform; Determine the real-time priority of each pending order data according to the real-time order processing status of each pending order data in the physical platform; According to the real-time priority of each logistics network point, the real-time priority of each salesperson and the real-time priority of each pending order data, the target network point and target salesperson for each pending order data are determined.

[0053] In the embodiments of this specification, the order priority is dynamically adjusted: a dynamic order priority adjustment mechanism based on real-time data feedback is established. By real-time monitoring of order processing, branch load, salesperson work status and other information, the order priority is adjusted in real time using a dynamic programming algorithm. For example, when the order backlog of a branch exceeds the set threshold, the priority of the branch's related order splitting algorithm is increased to ensure that the order can be quickly assigned to the appropriate processing link.

[0054] Requirement description: a. Request the map interface based on the order address to obtain the community name corresponding to the order (package pile name, package pile code, group name, group code, terminal code) b. New output fields: package pile name, package pile code, group name, group code, terminal code; c. Provide salesmen with a query interface for the collection and dispatch system, so that they can aggregate orders by community; Order monitoring report 1. Order splitting algorithm monitoring report a. Demand background: b. The current order splitting is a serial calculation, that is, once a certain order splitting algorithm has a result, the order splitting process is terminated; c. Order splitting priority is an internal setting of the business platform, and the code needs to be changed every time it is adjusted; 2. Demand pain points: a. Report statistics: When calculating report statistics, different order splitting methods have different order splitting quantities. Generally, the order splitting methods with higher priorities have higher order splitting proportions and higher accuracy. If the result cannot be calculated, the calculation will be carried out by the next priority. b. Optimal output: It is impossible to fully utilize the advantages of each algorithm to configure the strategy; 3. Purpose: For each order, all order splitting algorithms should output results. And the order splitting results of each algorithm should be recorded in the full order splitting result queue for report statistics. 4. Requirement description: a. The priority of the order splitting algorithm is changed to parallel calling; that is, each order splitting method is called, and then the priority of each order splitting method is set, and the results are output according to the priority; if a certain order splitting result times out (timeout setting time: xxx) or the returned result is empty, the order splitting method will not be included in the output; b. Order splitting result record: save the full order splitting results and record the results of each order splitting algorithm: i. The fields are as follows: 1. Order number: Take the waybill number / partner order number / order number; according to the existing rules, take the global universal value; to facilitate problem tracing and location. 2. Order splitting algorithm 1: current enumeration value; 3. Order splitting algorithm 1-predicted outlets: automatic order splitting and output of outlet codes 4. Order splitting algorithm 1-predicted pickup code: automatically split the order and output the pickup code. If there is no pickup code, this field will be empty. 5. Order splitting algorithm 1-predicted salesperson: the salesperson queried based on the outlet + pickup code or the salesperson directly returned by the interface; 6. Order splitting algorithm 2: 7. Order splitting algorithm 2-predicted outlets: automatic order splitting and output of outlet codes 8. Order splitting algorithm 2-predicted pickup code: Automatically split the order and output the pickup code. If there is no pickup code, this field will be empty. 9. Order splitting algorithm 2-predicted salesperson: the salesperson queried based on the outlet + pickup code or the interface directly Returning salesperson; 10. Order splitting algorithm N: 11. Order splitting algorithm N-prediction outlets: automatic order splitting and output of outlet codes 12. Order splitting algorithm N-predicted pickup code: Automatically split the order and output the pickup code. If there is no pickup code, this field will be empty. 13. Order splitting algorithm N-predicted salesperson: the salesperson queried based on the outlet + pickup code or the salesperson directly returned by the interface; c. Add a new terminal code splitting result queue: i. The fields are as follows: 1. Order number: 2. Predicted outlet: The outlet corresponding to the pickup code; 3. Predicted pickup code: This pickup code is a value generated by the relationship between the package pile and the pickup code; 4. Package pile name 5. Package stack coding 6. Group name 7. Terminal code 8. Order splitting algorithm: 9. Order calculation type: full word, keyword, etc.; 10. Calculation basis for splitting orders: address, keywords, longitude and latitude.

[0055] In the embodiment of this specification, the method further includes: Obtaining logistics order data whose historical order splitting results are erroneous results from the plurality of historical logistics order data to obtain wrongly split logistics order data; Input the misclassified logistics order data and the order splitting results into a machine learning model to extract order features, order splitting algorithm execution features, and external environment features; Identify the results of misclassification factors based on the order characteristics, order splitting algorithm execution characteristics, and external environment characteristics; According to the difference between the misclassification factor result and the real misclassification factor of the misclassified logistics order data, the machine learning model is trained to obtain a misclassification cause analysis model; Inputting the pending order data and the real-time order splitting results into the misclassification cause analysis model to perform misclassification factor analysis and obtain target misclassification factors; According to the target misclassification factor, an early warning prompt message is generated or the real-time order splitting result is adjusted according to the target misclassification factor to generate a revised order splitting result.

[0056] In the embodiments of this specification, loss information can be determined based on the difference between the misclassification factor result and the real misclassification factor of the misclassified logistics order data, and the parameters of the machine learning model can be adjusted according to the loss information until the training end conditions are met, and the machine learning model at the end of the training is determined as the misclassification cause analysis model. Therefore, the misclassification factor can be predicted according to the model, and early warning prompt information can be generated for the target misclassification factor, or the real-time order splitting result can be adjusted according to the target misclassification factor to generate a corrected order splitting result. The corrected order splitting result can also be displayed to the user, realizing the intelligent correction of the order splitting result.

[0057] In the embodiments of this specification, the calculation basis is recorded and traced as follows: Design a complete data recording and tracing system to record the calculation basis of each node in the order splitting process for each order. Use the tamper-proof characteristics of blockchain technology to ensure the authenticity and reliability of the data. For example, when address splitting is used, the calculation basis such as the order number, learned salesperson information, and address characteristics are recorded in detail; when using the pickup code keyword splitting, record the keywords, predicted outlets, pickup codes and other information, so that when problems arise, the source of the order splitting decision can be quickly traced back.

[0058] Misclassification cause analysis model: Build a misclassification cause analysis model based on big data analysis and machine learning. The model collects historical misclassified order data, analyzes order characteristics, order splitting algorithm execution, external environmental factors and other information, and uses the convolutional neural network model in the deep learning algorithm to automatically identify misclassification patterns, predict factors that may cause misclassification, and provide corresponding improvement suggestions. For example, by analyzing a large number of orders that were misclassified due to scheduling settings, the model can learn the correlation pattern between scheduling rules and misclassification, so as to provide early warning or automatic adjustment in the subsequent order splitting process.

[0059] This technical solution provides an order processing method based on an intelligent order splitting algorithm, through calculation basis recording and tracing: design a complete data recording and tracing system to record the calculation basis of each node in the order splitting process for each order. The tamper-proof characteristics of blockchain technology are used to ensure the authenticity and reliability of the data. For example, when address splitting is adopted, the calculation basis such as order number, learned salesperson information and address characteristics are recorded in detail; when using the pickup code keyword to split the order, the keywords, predicted outlets and pickup codes and other information are recorded, so that the source of the order splitting decision can be quickly traced when problems arise. Misclassification cause analysis model: Construct a misclassification cause analysis model based on big data analysis and machine learning. The model collects historical misclassified order data, analyzes order characteristics, order splitting algorithm execution, external environmental factors and other information, and uses the convolutional neural network model in the deep learning algorithm to automatically identify misclassification patterns, predict factors that may cause misclassification, and provide corresponding improvement suggestions. For example, by analyzing a large number of orders that are misclassified due to scheduling settings, the model can learn the correlation pattern between scheduling rules and misclassification, so as to provide early warning or automatic adjustment in the subsequent order splitting process. It can more accurately allocate orders to appropriate outlets, salespersons and pickup codes based on order characteristics, reducing the occurrence of misclassification.

[0060] In the embodiments of this specification, through precise data cleaning and standardization processing, as well as the integration of multiple algorithms and dynamic priority adjustment in the intelligent order splitting model, it is possible to more accurately allocate orders to appropriate outlets, salespersons and pickup codes based on order characteristics, reduce the occurrence of misclassification, and improve the accuracy of automatic order splitting.

[0061] The misclassification cause analysis model in the data tracing and analysis module can continuously learn and optimize, further improving the accuracy of order sorting and reducing the misclassification rate.

[0062] The multi-algorithm fusion order splitting and dynamic priority adjustment mechanism make the order splitting process more flexible and efficient, and can optimize the order splitting strategy in real time according to actual operating conditions to improve order processing efficiency.

[0063] The data traceability function helps to quickly locate problems in the order splitting process, make timely adjustments and optimizations, and reduce process delays caused by order splitting errors.

[0064] The complete data recording and traceability system provides transparent data support for logistics operations, making it convenient for managers and customers to inquire about the detailed process and basis of order splitting at any time.

[0065] The analysis results based on data tracing can be displayed through visual reports, which can intuitively present key information such as the order splitting effect and the reasons for mis-sorting, providing a strong basis for decision-making.

[0066] It can be seen from the technical solutions provided in the above embodiments of this specification that the embodiments of this specification obtain historical logistics order data and determine the historical order splitting results of the historical logistics order data; input the historical logistics order data into a preset model, and respectively use the address clerk order splitting algorithm, the pickup code electronic fence algorithm and the branch fixed clerk algorithm to split the historical logistics order data to obtain a first order splitting result corresponding to the address clerk order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm and a third order splitting result corresponding to the branch fixed clerk algorithm; obtain a first order splitting weight corresponding to the address clerk order splitting algorithm, a second order splitting weight corresponding to the pickup code electronic fence algorithm and a third order splitting result corresponding to the branch fixed clerk algorithm. The third order splitting weight corresponding to the fixed salesperson algorithm; determine the predicted order splitting result according to the first product of the first order splitting result and the first order splitting weight, the second product of the second order splitting result and the second order splitting weight, and the third product of the third order splitting result and the third order splitting weight; train the preset model according to the difference between the predicted order splitting result and the historical order splitting result to obtain an intelligent order splitting model; obtain the pending logistics order, input the pending logistics order into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting result; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the fixed salesperson algorithm of the outlet. The present invention can quickly trace back to the source of the order splitting decision, can quickly trace back to the source of the order splitting decision, and realize the intelligent, accurate, and rapid order splitting of logistics orders.

[0067] The embodiment of this specification also provides an intelligent order splitting device for logistics orders, such as Figure 6 As shown, the device comprises: A historical order splitting result determination module 610 is used to obtain historical logistics order data and determine the historical order splitting results of the historical logistics order data; The order splitting result determination module 620 is used to input the historical logistics order data into a preset model, and respectively adopt the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the network point fixed salesperson algorithm to split the historical logistics order data, and obtain a first order splitting result corresponding to the address salesperson order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the network point fixed salesperson algorithm; The order splitting weight determination module 630 is used to obtain the first order splitting weight corresponding to the address salesperson order splitting algorithm, the second order splitting weight corresponding to the pickup code electronic fence algorithm, and the third order splitting weight corresponding to the network point fixed salesperson algorithm; An order splitting result prediction module 640 is used to determine a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; An intelligent order splitting model training module 650 is used to train the preset model according to the difference between the predicted order splitting result and the historical order splitting result to obtain an intelligent order splitting model; The order splitting processing module 660 is used to obtain the logistics orders to be processed, input the logistics orders to be processed into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting results; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm.

[0068] In an exemplary embodiment, the historical order splitting result determination module includes: A data acquisition unit, used to acquire initial logistics order data; A cleaning unit, used to clean the initial logistics order data using a data cleaning algorithm to remove noise and error information to obtain cleaned order data; The data unification unit is used to convert the address, customer information, and product information in the cleaned order data into a unified format using a standardized algorithm to obtain the historical logistics order data.

[0069] In an exemplary embodiment, the apparatus further comprises: A historical information extraction module, used to extract the historical order number and historical address information of the historical logistics order data by adopting an address splitting strategy; A first database construction module is used to obtain the historical pickup salesperson corresponding to the historical order number, and to construct a corresponding relationship between the historical address information and the historical pickup salesperson to form a first database; A second database construction module is used to extract historical keywords, historical outlets and historical pickup codes of the historical logistics order data using a pickup code keyword strategy, and to construct a corresponding relationship between the historical keywords and the historical outlets and the historical pickup codes to form a second database; An electronic fence determination module, used to extract the historical longitude and latitude, historical outlets and historical pickup codes of the historical logistics order data by adopting the pickup code electronic fence strategy, and determine the historical electronic fence corresponding to the historical longitude and latitude; The third database construction module is used to construct the correspondence between the historical electronic fences and the historical outlets and the historical collection codes to form a third database.

[0070] In an exemplary embodiment, the apparatus further comprises: A first information acquisition module is used to acquire the real-time address information of the order data to be processed, and to search the first database for a pickup agent corresponding to the real-time address information to obtain a target pickup agent; or A second information acquisition module is used to acquire the real-time keywords of the order data to be processed, and search the second database for the real-time outlets and real-time pickup codes corresponding to the real-time keywords; or A third information acquisition module is used to obtain the real-time longitude and latitude of the pending order data and determine the real-time electronic fence corresponding to the real-time longitude and latitude; The data search module is used to search the real-time network point and the real-time pickup code corresponding to the real-time electronic fence in the third database.

[0071] In an exemplary embodiment, the order splitting result prediction module includes: A first calculation unit, used for calculating a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; The second calculation unit is used to calculate the sum of the first product, the second product and the third product to obtain the predicted order splitting result.

[0072] In an exemplary embodiment, the apparatus further comprises: A first priority determination module is used to determine the real-time priority of each logistics outlet according to the real-time order backlog of each logistics outlet in the logistics platform; A second priority determination module is used to determine the real-time priority of each salesperson according to the real-time working status of each salesperson in the logistics platform; A third priority determination module, configured to determine the real-time priority of each to-be-processed order data according to the real-time order processing status of each to-be-processed order data in the physical platform; The data determination module is used to determine the target outlet and target salesperson to which each pending order data is assigned based on the real-time priority of each logistics outlet, the real-time priority of each salesperson, and the real-time priority of each pending order data.

[0073] In an exemplary embodiment, the apparatus further comprises: A wrongly sorted data acquisition module is used to acquire logistics order data whose historical order splitting results are wrong results from the plurality of historical logistics order data to obtain the wrongly sorted logistics order data; A feature extraction module is used to input the misclassified logistics order data and the order splitting results into a machine learning model to extract order features, order splitting algorithm execution features and external environment features; A result identification module, used to identify the result of misclassification factors according to the order characteristics, the execution characteristics of the order splitting algorithm and the external environment characteristics; A model training module, used to train the machine learning model according to the difference between the misclassification factor result and the real misclassification factor of the misclassified logistics order data, so as to obtain a misclassification cause analysis model; A factor determination module, used for inputting the to-be-processed order data and the real-time order splitting results into the misclassification cause analysis model to perform misclassification factor analysis and obtain target misclassification factors; The correction module is used to generate early warning prompt information according to the target misclassification factor or adjust the real-time order splitting result according to the target misclassification factor to generate a corrected order splitting result.

[0074] The device and method embodiments in the described device embodiments are based on the same inventive concept.

[0075] An embodiment of the present specification provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement an intelligent order splitting method for logistics orders as provided in the above method embodiment.

[0076] An embodiment of the present invention also provides a computer storage medium, which can be set in a terminal to store at least one instruction or at least one program related to a method for implementing an intelligent order splitting method for logistics orders in a method embodiment. The at least one instruction or at least one program is loaded and executed by the processor to implement the intelligent order splitting method for logistics orders provided in the above method embodiment.

[0077] The embodiment of the present invention further provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes to implement the intelligent order splitting method for logistics orders provided in the above method embodiment.

[0078] Optionally, in the embodiments of this specification, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0079] The memory described in the embodiments of this specification can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, application programs required for functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0080] The intelligent order splitting method embodiment of the logistics order provided in the embodiments of this specification can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Taking running on a server as an example, Figure 7 This is a hardware structure block diagram of a server of an intelligent order splitting method for logistics orders provided in an embodiment of this specification. Figure 7As shown, the server 700 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 710 (the central processing unit 710 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 730 for storing data, and one or more storage media 720 (such as one or more mass storage devices) for storing application programs 723 or data 722. Among them, the memory 730 and the storage medium 720 can be short-term storage or permanent storage. The program stored in the storage medium 720 may include one or more modules, each of which may include a series of instruction operations on the server. Furthermore, the central processing unit 710 can be configured to communicate with the storage medium 720 and execute a series of instruction operations in the storage medium 720 on the server 700. The server 700 may also include one or more power supplies 760, one or more wired or wireless network interfaces 750, one or more input and output interfaces 740, and / or one or more operating systems 721, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0081] The input / output interface 740 may be used to receive or send data via a network. A specific example of the network may include a wireless network provided by a communication provider of the server 700. In one example, the input / output interface 740 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 740 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0082] It can be understood by those skilled in the art that Figure 7 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 7 More or fewer components as shown, or with Figure 7 Different configurations are shown.

[0083] It can be seen from the embodiments of the intelligent order splitting method, device, electronic device or storage medium for logistics orders provided by the present invention that the present invention obtains historical logistics order data and determines the historical order splitting results of the historical logistics order data; inputs the historical logistics order data into a preset model, and respectively uses the address clerk order splitting algorithm, the pickup code electronic fence algorithm and the branch fixed clerk algorithm to split the historical logistics order data to obtain a first order splitting result corresponding to the address clerk order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm and a third order splitting result corresponding to the branch fixed clerk algorithm; obtains a first order splitting weight corresponding to the address clerk order splitting algorithm, a second order splitting weight corresponding to the pickup code electronic fence algorithm Weight and the third order splitting weight corresponding to the algorithm of the fixed salesperson at the outlet; determine the predicted order splitting result according to the first product of the first order splitting result and the first order splitting weight, the second product of the second order splitting result and the second order splitting weight, and the third product of the third order splitting result and the third order splitting weight; train the preset model according to the difference between the predicted order splitting result and the historical order splitting result to obtain an intelligent order splitting model; obtain the pending logistics order, input the pending logistics order into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting result; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the fixed salesperson algorithm at the outlet. The present invention can quickly trace back to the source of the order splitting decision, can quickly trace back to the source of the order splitting decision, and realize the intelligent, accurate and fast order splitting of logistics orders.

[0084] It should be noted that the above sequence of the embodiments of this specification is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0086] A person skilled in the art will appreciate that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent order splitting method for logistics orders, characterized in that: The method comprises: Obtaining historical logistics order data, and determining historical order splitting results of the historical logistics order data; The historical logistics order data is input into a preset model, and the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the network point fixed salesperson algorithm are respectively used to split the historical logistics order data to obtain a first order splitting result corresponding to the address salesperson order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the network point fixed salesperson algorithm; Obtain a first order distribution weight corresponding to the address salesperson order distribution algorithm, a second order distribution weight corresponding to the pickup code electronic fence algorithm, and a third order distribution weight corresponding to the network point fixed salesperson algorithm; Determine a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; According to the difference between the predicted order splitting result and the historical order splitting result, the preset model is trained to obtain an intelligent order splitting model; Obtain the pending logistics orders, input the pending logistics orders into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting results; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm.

2. The method according to claim 1, characterized in that: The obtaining of historical logistics order data includes: Get initial logistics order data; Using a data cleaning algorithm to clean the initial logistics order data, remove noise and erroneous information, and obtain cleaned order data; A standardized algorithm is used to convert the address, customer information, and product information in the cleaned order data into a unified format to obtain the historical logistics order data.

3. The method according to claim 1, characterized in that The method further comprises: Adopting the address splitting strategy to extract the historical order number and historical address information of the historical logistics order data; Obtaining the historical pickup clerk corresponding to the historical order number, and constructing a corresponding relationship between the historical address information and the historical pickup clerk to form a first database; The historical keywords, historical outlets, and historical pickup codes of the historical logistics order data are extracted using a pickup code keyword strategy, and a corresponding relationship between the historical keywords and the historical outlets and the historical pickup codes is constructed to form a second database; The pickup code electronic fence strategy is adopted to extract the historical longitude and latitude, historical outlets and historical pickup codes of the historical logistics order data, and determine the historical electronic fence corresponding to the historical longitude and latitude; A correspondence between the historical electronic fences, the historical outlets, and the historical pickup codes is constructed to form a third database.

4. The method according to claim 3, characterized in that The method further comprises: Acquire the real-time address information of the order data to be processed, search the first database for a pickup agent corresponding to the real-time address information, and obtain a target pickup agent; or Obtaining a real-time keyword of the order data to be processed, and searching the second database for a real-time outlet and a real-time pickup code corresponding to the real-time keyword; or Obtaining the real-time longitude and latitude of the pending order data, and determining the real-time electronic fence corresponding to the real-time longitude and latitude; The real-time network point and the real-time pickup code corresponding to the real-time electronic fence are searched in the third database.

5. The method according to claim 1, characterized in that The step of determining a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight includes: Calculate a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; The sum of the first product, the second product and the third product is calculated to obtain the predicted order splitting result.

6. The method according to claim 1, characterized in that The method further comprises: Determine the real-time priority of each logistics outlet based on the real-time order backlog of each logistics outlet in the logistics platform; Determine the real-time priority of each salesperson according to the real-time working status of each salesperson in the logistics platform; Determine the real-time priority of each pending order data according to the real-time order processing status of each pending order data in the physical platform; According to the real-time priority of each logistics network point, the real-time priority of each salesperson and the real-time priority of each pending order data, the target network point and target salesperson for each pending order data are determined.

7. The method according to claim 1, characterized in that The method further comprises: Obtaining logistics order data whose historical order splitting results are erroneous results from the plurality of historical logistics order data to obtain wrongly split logistics order data; Input the misclassified logistics order data and the order splitting results into a machine learning model to extract order features, order splitting algorithm execution features, and external environment features; Identify the results of misclassification factors based on the order characteristics, order splitting algorithm execution characteristics, and external environment characteristics; According to the difference between the misclassification factor result and the real misclassification factor of the misclassified logistics order data, the machine learning model is trained to obtain a misclassification cause analysis model; Inputting the pending order data and the real-time order splitting results into the misclassification cause analysis model to perform misclassification factor analysis and obtain target misclassification factors; According to the target misclassification factor, an early warning prompt message is generated or the real-time order splitting result is adjusted according to the target misclassification factor to generate a revised order splitting result.

8. An intelligent order splitting device for logistics orders, characterized in that: The device comprises: A historical order splitting result determination module is used to obtain historical logistics order data and determine the historical order splitting results of the historical logistics order data; An order splitting result determination module is used to input the historical logistics order data into a preset model, and respectively adopt an address salesperson order splitting algorithm, a pickup code electronic fence algorithm, and a network fixed salesperson algorithm to split the historical logistics order data, and obtain a first order splitting result corresponding to the address salesperson order splitting algorithm, a second order splitting result corresponding to the pickup code electronic fence algorithm, and a third order splitting result corresponding to the network fixed salesperson algorithm; An order splitting weight determination module is used to obtain a first order splitting weight corresponding to the address salesperson order splitting algorithm, a second order splitting weight corresponding to the pickup code electronic fence algorithm, and a third order splitting weight corresponding to the fixed salesperson algorithm at the outlet; An order splitting result prediction module, used to determine a predicted order splitting result according to a first product of the first order splitting result and the first order splitting weight, a second product of the second order splitting result and the second order splitting weight, and a third product of the third order splitting result and the third order splitting weight; An intelligent order splitting model training module, used to train the preset model according to the difference between the predicted order splitting result and the historical order splitting result to obtain an intelligent order splitting model; The order splitting processing module is used to obtain the logistics orders to be processed, input the logistics orders to be processed into the intelligent order splitting model for order splitting processing, and obtain the target order splitting algorithm and order splitting results; the target order splitting algorithm is at least one of the address salesperson order splitting algorithm, the pickup code electronic fence algorithm, and the branch fixed salesperson algorithm.

9. An electronic device, characterized in that: The device includes: a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the intelligent order splitting method for logistics orders as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the intelligent order splitting method for logistics orders as described in any one of claims 1-7.

Images