Logistics distribution management method, device and system

Abstract

The invention discloses a logistics distribution management method, device and system, and relates to the field of logistics management. The method comprises the steps that a distribution prediction request for a logistics line is received, and the distribution prediction request comprises a line identifier and a prediction time period of the logistics line; dynamic feature data and static feature data related to the line identifier and the prediction time period are obtained, and the dynamic feature data comprise order features, warehouse production features and order sorting features related to online real-time article orders; and utilizing a pre-trained prediction model, the dynamic characteristic data and the static characteristic data to predict logistics distribution information which corresponds to the prediction time period and indicates the departure shift and the departure frequency. According to the scheme provided by the embodiment of the invention, the logistics distribution management flexibility can be effectively improved, and the logistics distribution efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of logistics management technology, and in particular to a logistics distribution management method, apparatus and system. Background Technology

[0002] Logistics distribution mainly refers to arranging the departure schedule and frequency of logistics routes based on the amount of goods to be transported, in order to control logistics transportation costs while avoiding the accumulation of goods as much as possible.

[0003] Currently, logistics distribution is primarily managed manually based on historical order data. For example, manual scheduling uses historical data and experience to determine departure times and frequencies for logistics routes. However, with the increasing prevalence of online shopping promotions, and the varying promotional periods for different product categories, the items requiring transportation become more complex and variable, directly impacting logistics operations such as departure times and frequencies. The existing manual management method, which maintains departure times and frequencies based solely on historical orders, does not comprehensively consider online shopping orders and the complex and ever-changing online promotional activities. Therefore, the current manual management method leaves room for improvement in terms of logistics distribution flexibility and efficiency. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a logistics distribution management method, apparatus and system that can effectively improve the flexibility of logistics distribution management and increase logistics distribution efficiency.

[0005] To achieve the above objectives, in a first aspect, embodiments of the present invention provide a logistics distribution management method, comprising:

[0006] Receive a delivery forecast request for a logistics route, wherein the delivery forecast request includes the route identifier of the logistics route and the forecast time period;

[0007] Acquire dynamic and static feature data related to the route identifier and the predicted time period, wherein the dynamic feature data includes order features, warehouse production features, and order sorting features related to online real-time item orders;

[0008] Using a pre-trained prediction model, the dynamic feature data, and the static feature data, logistics distribution information corresponding to the indicated departure times and departure frequencies for the predicted time period is predicted.

[0009] Optionally,

[0010] The acquisition of dynamic feature data related to the line identifier and the predicted time period includes:

[0011] Based on the sending and receiving addresses of online real-time item orders generated within a set time period, target online real-time item orders that match the route identifier are filtered out, and based on the item information included in the target online real-time item orders, the package quantity, package volume, and logistics waybill volume are predicted.

[0012] Obtain real-time production data of the warehouse used to produce the online real-time item orders, and extract warehouse feature values ​​corresponding to the warehouse production features from the real-time production data;

[0013] Obtain order sorting data for the sorting process used to process the online real-time item orders, and extract the sorting feature values ​​corresponding to the order sorting features from the order sorting data.

[0014] Optionally, static feature data related to the line identifier and the predicted time period are obtained, including:

[0015] Based on the predicted time period, determine the historical time period;

[0016] Based on the logistics and distribution data corresponding to the historical time period, analyze the route characteristics and route profiles corresponding to the route identifier.

[0017] Optionally, the above-mentioned logistics distribution management method also includes:

[0018] Based on the route, online historical item orders are grouped according to historical package volume, historical package size, and historical waybill volume;

[0019] According to the route, extract the warehouse historical feature value corresponding to the warehouse production feature from the warehouse historical production data, and extract the sorting historical feature value corresponding to the order sorting feature from the order sorting historical data;

[0020] Using the grouping results, the warehouse's historical feature values, the sorting's historical feature values, and the route information, the model containing the decision tree is iteratively trained to obtain the prediction model.

[0021] Optionally, the above-mentioned logistics distribution management method also includes:

[0022] For each iteration of the iterative training of the model containing the decision tree, perform the following operation:

[0023] Adjust the warehouse production characteristics and / or the order sorting characteristics, and use the grouping results and the warehouse historical feature values ​​corresponding to the adjusted warehouse production characteristics and / or the sorting historical feature values ​​corresponding to the adjusted order sorting characteristics to train a model containing decision trees;

[0024] Analyze the correlation between the adjusted warehouse production characteristics and the adjusted order sorting characteristics and logistics distribution management;

[0025] Based on the analysis results, the warehouse production characteristics and / or order sorting characteristics used to train the model are revised.

[0026] Optionally, the line characteristics include any one or more of the following characteristics:

[0027] Data distribution of route type, relationships between routes, historical transport volume, historical train number, and historical load factor;

[0028] And / or,

[0029] The route profile includes the operational status and profit and loss information corresponding to the route identifier.

[0030] Optionally, the prediction corresponds to logistics distribution information including indicated departure times and frequency for the predicted time period, including:

[0031] The pre-trained prediction model uses the dynamic feature data and the static feature data, including the warehouse production features and the order sorting features, to analyze multiple departure shifts, multiple departure frequencies and vehicle loads, and predicts the probability corresponding to each departure shift, departure frequency and vehicle load.

[0032] Optionally, the prediction of logistics distribution information corresponding to the indicated departure times and frequencies for the predicted time period further includes:

[0033] Filter and output the target departure times, target departure frequencies, and target vehicle loads corresponding to predicted probabilities that are not less than a preset threshold.

[0034] Secondly, embodiments of the present invention provide a logistics distribution management device, comprising: an interaction module and a predictive processing module, wherein...

[0035] The interaction module is used to receive a delivery forecast request for a logistics route, wherein the delivery forecast request includes the route identifier of the logistics route and the forecast time period.

[0036] The prediction processing module is used to acquire dynamic feature data and static feature data related to the route identifier and the prediction time period. The dynamic feature data includes order features, warehouse production features and order sorting features related to online real-time item orders. Using a pre-trained prediction model, the dynamic feature data and the static feature data, the module predicts the logistics distribution information corresponding to the indicated departure times and departure frequencies for the prediction time period.

[0037] Thirdly, embodiments of the present invention provide a logistics distribution management system, comprising: an online order management terminal and the logistics distribution management device provided in the second aspect of the embodiments described above, wherein...

[0038] The online order management terminal is used to provide the logistics distribution management device with dynamic data related to orders;

[0039] The logistics distribution management device is further used to acquire dynamic feature data related to route identification and predicted time period based on dynamic data related to orders, wherein the dynamic feature data includes order features, warehouse production features and order sorting features related to online real-time item orders.

[0040] One embodiment of the above invention has the following advantages or beneficial effects: After receiving a delivery prediction request including the route identifier and predicted time period of the logistics route, the system uses a pre-trained prediction model and acquired dynamic and static feature data related to the route identifier and predicted time period, including order features, warehouse production features, and order sorting features related to online real-time item orders, to predict the logistics delivery information corresponding to the indicated departure schedule and departure frequency for the predicted time period. Since dynamic feature data including order features, warehouse production features, and order sorting features related to online real-time item orders are introduced in the prediction process, and since logistics delivery is mainly for items sold on online e-commerce platforms, the solution provided in this application can effectively improve the flexibility of logistics delivery management and increase logistics delivery efficiency by associating departure schedules and departure frequencies with orders.

[0041] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0042] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0043] Figure 1 This is a schematic diagram of the main process of a logistics distribution management method according to an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the main process for obtaining dynamic and static feature data related to line identification and predicted time period according to an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the main process of training a model in a logistics distribution management method according to another embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram of the main process of a logistics distribution management method according to another embodiment of the present invention;

[0047] Figure 5 This is a schematic diagram of the main modules of a logistics distribution management device according to an embodiment of the present invention;

[0048] Figure 6 This is a schematic diagram of the main equipment of the logistics distribution management system according to an embodiment of the present invention;

[0049] Figure 7 This is an exemplary system architecture diagram in which embodiments of the present invention can be applied;

[0050] Figure 8 This is a schematic diagram of the structure of a computer system suitable for implementing logistics distribution management equipment or terminal equipment in the embodiments of the present invention. Detailed Implementation

[0051] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0052] Figure 1 This is a schematic diagram of the main process of a logistics distribution management method according to an embodiment of the present invention. Figure 1 As shown, this logistics distribution management method may include the following steps:

[0053] Step S101: Receive a shipment forecast request for a logistics route, wherein the shipment forecast request includes the route identifier of the logistics route and the forecast time period.

[0054] In general, the route identification of a logistics route refers to a unique code or number that identifies the route.

[0055] The predicted time period generally refers to a future time, such as a certain day in the future, a week in the future starting from the current time, or five days in the future starting from the current time.

[0056] Logistics routes can include trunk lines, such as long-distance transportation between sorting centers or transfer centers across provinces; branch lines, such as long-distance transportation between sorting centers or transfer centers that do not cross provinces but cross cities; transfer stations, such as short-distance transportation from sorting centers or transfer centers in the same city to business departments or distribution stations; and shuttle transportation, such as short-distance transportation from warehouses in the same city to sorting centers or transfer centers.

[0057] Step S102: Obtain dynamic and static feature data related to the route identifier and the predicted time period. The dynamic feature data includes order features, warehouse production features, and order sorting features related to online real-time item orders.

[0058] Among them, dynamic feature data is generally related to the order information generated by online e-commerce platforms. This dynamic feature data is constantly changing and can intuitively reflect the situation of the items that need to be shipped along the route, such as the quantity and volume of items. This dynamic feature data generally comes from orders generated in real time by the e-commerce platform's order management system based on user order information, the quantity and volume of packages produced by the warehouse in the warehouse management system, and the sorting efficiency and sorting volume in the sorting process.

[0059] Step S103: Using a pre-trained prediction model, dynamic feature data, and static feature data, predict the logistics distribution information corresponding to the indicated departure times and departure frequencies for the predicted time period.

[0060] The number of departures generally refers to the number of vehicles that depart on a route within a day, while the departure frequency generally refers to the interval between two adjacent departures on a route.

[0061] exist Figure 1 In the illustrated embodiment, upon receiving a delivery prediction request including the route identifier and predicted time period of the logistics route, the system uses a pre-trained prediction model and acquired dynamic and static feature data related to the route identifier and predicted time period, including order features, warehouse production features, and order sorting features related to online real-time item orders, to predict the logistics delivery information corresponding to the indicated departure schedule and departure frequency for the predicted time period. Since dynamic feature data including order features, warehouse production features, and order sorting features related to online real-time item orders are introduced during the prediction process, and since logistics delivery is mainly for items sold through online e-commerce, the solution provided in this application can effectively improve the flexibility of logistics delivery management and increase logistics delivery efficiency by associating departure schedules and departure frequencies with orders.

[0062] Specifically, such as Figure 2 As shown, a specific implementation of step S102 above may include the following steps:

[0063] Step S201: Based on the sending and receiving addresses of online real-time item orders generated within a set time period, filter out target online real-time item orders that match the route identifier, and predict the package quantity, package volume, and logistics waybill volume based on the item information included in the target online real-time item orders.

[0064] Generally, there are one or more logistics routes from the sender's address to the recipient's address. For example, for an order from the sender's address to the recipient's address, the route is from warehouse A to sorting center B, from sorting center B to transit station C, and from transit station C to distribution station D. Therefore, warehouse A to sorting center B is one logistics route, sorting center B to transit station C is another logistics route, and transit station C to distribution station D is yet another logistics route.

[0065] Step S202: Obtain real-time production data of the warehouse used for real-time item orders on the production line, and extract warehouse feature values ​​corresponding to warehouse production features from the real-time production data;

[0066] For example, the quantity of items leaving the warehouse, the quantity of packages, the volume of packages, the status of warehouse production processes such as whether there are any delays, warehouse staff schedules, and the number of staff on duty.

[0067] Step S203: Obtain order sorting data for the sorting process used to process online real-time item orders, and extract the sorting feature values ​​corresponding to the order sorting features from the order sorting data.

[0068] The efficiency of the sorting process, the scheduling of sorting personnel, and the adjustment of sorting operations, etc.

[0069] This invention utilizes data from e-commerce platforms, warehouse production status, sorting status, and logistics network traffic to mine real-time forecast order volume, which has a strong correlation with predictions. This data assists in judging the final departure status of bus routes and helps improve the accuracy of prediction results.

[0070] In addition, by introducing upstream information into the logistics distribution management process and selecting features and information that can have a significant impact on logistics distribution management from the upstream information (predicting parcel volume, parcel size, and waybill volume, and obtaining warehouse feature values ​​corresponding to warehouse production features and sorting feature values ​​corresponding to order sorting features), the rationality of logistics distribution management can be effectively improved. For example, it can avoid order backlog, avoid low loading rates of transport vehicles, and effectively control logistics distribution costs.

[0071] Furthermore, specific implementation methods for obtaining static feature data related to route identifiers and predicted time periods may include: determining historical time periods based on the predicted time periods; and analyzing the route features and route profiles corresponding to the route identifiers based on the logistics and distribution data corresponding to the historical time periods.

[0072] By predicting time periods and determining historical time periods, promotional periods and idle periods can be effectively divided. Based on historical time periods, it can be determined whether the predicted time period is a promotional period or an idle period. The historical time periods determined by the predicted time period generally match the dynamic feature data obtained above. By analyzing the logistics distribution data corresponding to the historical time periods, the route characteristics and route profiles corresponding to the route identifiers are analyzed. Combined with the dynamic feature data, the rationality of the logistics distribution information of the predicted departure times and frequencies and the accuracy of the prediction results can be further improved.

[0073] The line characteristics may include any one or more of the following characteristics:

[0074] Data distribution of route type, relationships between routes, historical transport volume, historical train number, and historical load factor.

[0075] Among them, the route types can be divided into four types: trunk, branch, transfer, and shuttle. For example, trunk (long-distance transportation between sorting centers or transfer centers across provinces), branch (long-distance transportation between sorting centers or transfer centers that do not cross provinces but cross cities), transfer (short-distance transportation from sorting centers or transfer centers in the same city to business departments or distribution stations), and shuttle (short-distance transportation from warehouses in the same city to sorting centers or transfer centers).

[0076] The relationships between routes: From address a to address b, it can be from sorting center 1 to transit station 2 (logistics route 1), from transit station 2 to transit station 3 (logistics route 2), and from transit station 3 to distribution station 4 (logistics route 3). Logistics route 1 is the upstream route of logistics route 2, and logistics route 3 is the upstream route of [the route].

[0077] The route profile can include the operational status and profit / loss information corresponding to the route identifier. By incorporating the operational and profit / loss information of logistics routes into the forecasting process, the cost of logistics routes can be effectively controlled.

[0078] Furthermore, such as Figure 3 As shown, the above-mentioned logistics distribution management method may further include the following steps:

[0079] Step S301: Group the historical parcel volume, historical parcel volume, and historical waybill volume of online historical item orders according to the route;

[0080] This step mainly involves grouping the historical parcel volume, historical parcel volume, and historical waybill volume belonging to the same logistics route into the same group. Historical parcel volume refers to the number of parcels transported by the logistics route within each historical time period; historical parcel volume refers to the volume of parcels transported by the logistics route within each historical time period; and historical waybill volume refers to the number of logistics waybills transported by the logistics route within each historical time period.

[0081] Step S302: According to the route, extract the warehouse historical feature value corresponding to the warehouse production feature from the warehouse historical production data, and extract the sorting historical feature value corresponding to the order sorting feature from the order sorting historical data;

[0082] The production characteristics and order sorting characteristics of this warehouse have already been described above and will not be repeated here.

[0083] Step S303: Using the grouping results, warehouse historical feature values, sorting historical feature values, and route information, iteratively train the model containing the decision tree to obtain the prediction model.

[0084] The route information used to train the model containing the decision tree includes departure times and frequency.

[0085] Furthermore, during training, the reliability of training the model containing the decision tree is improved by adjusting parameters such as the number of leaf nodes, the depth of the tree, and the learning rate, thereby increasing the accuracy of the prediction model. Specifically, in this embodiment of the invention, the depth of the decision tree is controlled to not exceed 6 and the number of leaves to not exceed 64 during training.

[0086] It is worth noting that the trained prediction model was exported in PMML format and deployed as a Java service.

[0087] By training the model using the historical parcel volume, historical parcel volume, and historical waybill volume of the logistics routes obtained above, as well as the warehouse production characteristics and order sorting characteristics, the accuracy of the prediction model training can be effectively improved.

[0088] Furthermore, the aforementioned logistics distribution management method may also include: for each iteration cycle of iteratively training the model containing the decision tree, performing the following operations:

[0089] Adjust warehouse production characteristics and / or order sorting characteristics, and use the grouping results and the warehouse historical feature values ​​corresponding to the adjusted warehouse production characteristics and / or the sorting historical feature values ​​corresponding to the adjusted order sorting characteristics to train a model containing decision trees;

[0090] Analyze the correlation between the adjusted warehouse production characteristics and the adjusted order sorting characteristics and logistics distribution management;

[0091] Based on the analysis results, the warehouse production characteristics and / or order sorting characteristics used to train the model are revised.

[0092] The above process corrects and adjusts warehouse production characteristics and / or order sorting characteristics to select characteristics that have a significant impact on the prediction results, thereby further improving the accuracy of the trained prediction model and the accuracy of the prediction results.

[0093] It is worth noting that the features in the dynamic feature data containing order features, warehouse production features, and order sorting features used in the above embodiments of the present invention, as well as the features in the static feature data, are obtained by combining the above adjustment and correction process.

[0094] Furthermore, the specific implementation method for the logistics distribution information corresponding to the predicted departure times and frequencies within the predicted time period may include: a pre-trained prediction model using dynamic and static feature data, including warehouse production characteristics and order sorting characteristics, to analyze multiple departure times, multiple departure frequencies, and vehicle loads, and predict the probability of each departure time, departure frequency, and vehicle load. This allows users to further adjust departure times and frequencies based on the departure times, departure frequencies, vehicle loads, and corresponding prediction probabilities, thereby further improving the rationality of logistics route management.

[0095] Furthermore, the aforementioned prediction of the indicated departure times and frequencies for the predicted time period can further include: filtering and outputting the target departure times, target departure frequencies, and target vehicle loads corresponding to prediction probabilities not less than a preset threshold. This process can further enhance the automation of logistics distribution management and reduce manual management.

[0096] The following section uses the example of training a model and using that model to predict logistics distribution information for a logistics route to explain the logistics distribution management method in detail. Specifically, for example... Figure 4 As shown, this logistics distribution management method may include the following steps:

[0097] Step S401: Obtain historical order data including order characteristics, warehouse production characteristics, and order sorting characteristics, as well as the relationships between routes, historical transport volume of routes, historical train trips, historical load rates, route operation status, and route profit and loss status.

[0098] Step S402: Using historical order data including order features, warehouse production features, and order sorting features, as well as the relationships between routes, historical transport volume of routes, historical train trips, historical loading rates, route operation status, and route profit and loss status, iteratively train a model containing decision trees, adjust the features used in the training model, and obtain a pre-trained prediction model.

[0099] Based on the route, online historical item orders are grouped according to historical package volume, historical package size, and historical waybill volume;

[0100] According to the route, extract the warehouse historical feature values ​​corresponding to the warehouse production characteristics from the warehouse historical production data, and extract the sorting historical feature values ​​corresponding to the order sorting characteristics from the order sorting historical data;

[0101] Using the grouping results, the warehouse's historical feature values, the sorting's historical feature values, and the route information, the model containing the decision tree is iteratively trained to obtain the prediction model.

[0102] Adjust the warehouse production characteristics and / or the order sorting characteristics, and use the grouping results and the warehouse historical feature values ​​corresponding to the adjusted warehouse production characteristics and / or the sorting historical feature values ​​corresponding to the adjusted order sorting characteristics to train a model containing decision trees;

[0103] Analyze the correlation between the adjusted warehouse production characteristics and the adjusted order sorting characteristics and logistics distribution management;

[0104] Based on the analysis results, the warehouse production characteristics and / or order sorting characteristics used to train the model are revised.

[0105] Step S403: Upon receiving a delivery forecast request for a logistics route, including the route identifier and forecast time period, obtain the forecasted parcel volume, parcel volume, and waybill volume related to the route identifier and forecast time period, warehouse feature values ​​corresponding to warehouse production characteristics, sorting feature values ​​corresponding to order sorting characteristics, route characteristics, and route profile.

[0106] Based on the sending and receiving addresses of online real-time item orders generated within a set time period, target online real-time item orders that match the route identifier are filtered out, and based on the item information included in the target online real-time item orders, the package quantity, package volume, and logistics waybill volume are predicted.

[0107] Acquire real-time production data of the warehouse for real-time item orders on the production line, and extract warehouse feature values ​​corresponding to warehouse production characteristics from the real-time production data;

[0108] Obtain order sorting data for the sorting process used to process the online real-time item orders, and extract sorting feature values ​​corresponding to the order sorting features from the order sorting data;

[0109] Based on the predicted time period, determine the historical time period;

[0110] Based on the logistics and distribution data corresponding to the historical time period, analyze the route characteristics and route profiles corresponding to the route identifier.

[0111] Step S404: Based on the predicted parcel volume, parcel size, and logistics waybill volume related to the route identifier and predicted time period, the warehouse feature value corresponding to the warehouse production characteristics, the sorting feature value corresponding to the order sorting characteristics, the route characteristics, and the route profile, the pre-trained prediction model, dynamic feature data, and static feature data are used to analyze multiple departure shifts, multiple departure frequencies, and vehicle load, and predict the prediction probability corresponding to each departure shift, departure frequency, and vehicle load.

[0112] Step S405: Filter and output the target departure times, target departure frequencies, and target vehicle loads corresponding to the predicted probabilities that are not less than the preset threshold.

[0113] Figure 5 This is a schematic diagram of the structure of a logistics distribution management device provided in an embodiment of the present invention. Figure 5 As shown, the logistics distribution management device 500 may include: an interaction module 501 and a predictive processing module 502, wherein,

[0114] The interaction module 501 is used to receive a delivery forecast request for a logistics route, wherein the delivery forecast request includes the route identifier of the logistics route and the forecast time period.

[0115] The prediction processing module 502 is used to acquire dynamic feature data and static feature data related to the route identifier and the prediction time period. The dynamic feature data includes order features, warehouse production features and order sorting features related to online real-time item orders. Using the pre-trained prediction model, dynamic feature data and static feature data, the module predicts the logistics distribution information corresponding to the indicated departure times and departure frequencies for the prediction time period.

[0116] In this embodiment of the invention, the prediction processing module 502 is further configured to: filter out target online real-time item orders that match the route identifier based on the sending address and receiving address of online real-time item orders generated within a set time period; and predict the package quantity, package volume, and logistics waybill quantity based on the item information included in the target online real-time item orders; acquire real-time production data of the warehouse used for real-time item orders on the production line, and extract warehouse feature values ​​corresponding to warehouse production characteristics from the real-time production data; acquire order sorting data of the sorting process used for processing online real-time item orders, and extract sorting feature values ​​corresponding to order sorting characteristics from the order sorting data.

[0117] In this embodiment of the invention, the prediction processing module 502 is further configured to determine the historical time period based on the prediction time period; and to analyze the route characteristics and route profile corresponding to the route identifier based on the logistics distribution data corresponding to the historical time period.

[0118] In this embodiment of the invention, the prediction processing module 502 is further configured to group the historical parcel volume, historical parcel volume, and historical waybill volume of online historical item orders according to the route; extract the warehouse historical feature value corresponding to the warehouse production feature from the warehouse historical production data according to the route, and extract the sorting historical feature value corresponding to the order sorting feature from the order sorting historical data; and use the grouping results, warehouse historical feature value, sorting historical feature value, and route information to iteratively train a model containing a decision tree to obtain a prediction model.

[0119] In this embodiment of the invention, the prediction processing module 502 is further configured to perform the following operations for each iteration of the iterative training of the model containing the decision tree:

[0120] Adjust warehouse production characteristics and / or order sorting characteristics; use the grouping results and the historical warehouse characteristic values ​​corresponding to the adjusted warehouse production characteristics and / or the historical sorting characteristic values ​​corresponding to the adjusted order sorting characteristics to train a model containing decision trees; analyze the correlation between the adjusted warehouse production characteristics and the adjusted order sorting characteristics and logistics distribution management; and based on the analysis results, revise the warehouse production characteristics and / or order sorting characteristics used to train the model.

[0121] In embodiments of the present invention, the line features include any one or more of the following features:

[0122] Data distribution of route type, relationships between routes, historical transport volume, historical train number, and historical load factor.

[0123] In this embodiment of the invention, the route profile includes the operational status and profit and loss status corresponding to the route identifier.

[0124] In this embodiment of the invention, the prediction processing module 502 is further used to analyze multiple departure shifts, multiple departure frequencies and vehicle loads using dynamic feature data and static feature data including warehouse production features and order sorting features of the pre-trained prediction model, and to predict the prediction probability corresponding to each departure shift, departure frequency and vehicle load.

[0125] In this embodiment of the invention, the prediction processing module 502 is further used to filter and output the target departure schedule, target departure frequency and target vehicle load corresponding to the prediction probability not less than a preset threshold.

[0126] This invention provides a logistics distribution management system. Figure 6 This diagram illustrates the structure of the logistics distribution management system. Figure 6 As shown, the logistics distribution management system 600 may include: an online order management terminal 601 and the logistics distribution management device 500 provided in the above embodiments, wherein,

[0127] The online order management terminal 601 is used to provide dynamic data related to orders to the logistics distribution management device 500;

[0128] The logistics distribution management device 500 is further used to acquire dynamic feature data related to route identification and predicted time period based on dynamic data related to orders. The dynamic feature data includes order features, warehouse production features and order sorting features related to online real-time item orders.

[0129] Figure 7 An exemplary system architecture 700 is shown that can be applied to the logistics distribution management method or logistics distribution management device of the present invention.

[0130] like Figure 7 As shown, the system architecture 700 may include a terminal device 701, a network 702, an e-commerce order management system 703, a logistics waybill management system 704, and a logistics distribution management device 705. The communication module 702 provides a communication medium between the terminal device 701 and the logistics distribution management device 705, between the e-commerce order management system 703 and the logistics distribution management device 705, and between the logistics waybill management system 704 and the logistics distribution management device 705. The network 702 may include various connection types, such as wired and wireless communication such as Bluetooth, infrared communication, and data cable communication.

[0131] The e-commerce order management system 703 is used to manage online orders placed by users, the production status of the warehouse for those orders, and sorting status.

[0132] The logistics distribution management device 705 can obtain dynamic feature data from the e-commerce order management system 703, including order features, warehouse production features, and order sorting features related to online real-time item orders, and static feature data from the logistics waybill management system 704, such as route features, route profiles, and historical waybill volumes for a certain historical period. Based on the obtained dynamic and static feature data, after receiving a distribution prediction request for a logistics route from the terminal device 701, it uses a pre-trained prediction model to predict the logistics distribution information of the indicated departure times and departure frequencies for the corresponding prediction period on the logistics route.

[0133] It should be noted that the logistics distribution management method provided in the embodiments of the present invention is generally completed by the logistics distribution management device 705. Accordingly, each module of the logistics distribution management device can be set in the logistics distribution management device 705.

[0134] It should be understood that Figure 7The number of terminal devices, networks, e-commerce order management systems, logistics waybill management systems, and logistics distribution management devices shown in the diagram is merely illustrative. Depending on the implementation requirements, any number of terminal devices, networks, e-commerce order management systems, logistics waybill management systems, and logistics distribution management devices can be included.

[0135] The following is for reference. Figure 8 It shows a schematic diagram of the structure of a computer system 800 suitable for implementing logistics distribution management equipment or terminal equipment in the embodiments of the present invention. Figure 8 The logistics distribution management equipment or terminal equipment shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0136] like Figure 8 As shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 802 or programs loaded from storage section 808 into random access memory (RAM) 803. The RAM 803 also stores various programs and data required for the operation of the system 800. The CPU 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0137] The following components are connected to I / O interface 805: an input section 806 including a keyboard, mouse, etc.; an output section 807 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to I / O interface 805 as needed. A removable medium 811, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 810 as needed so that computer programs read from it can be installed into storage section 808 as needed.

[0138] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 809, and / or installed from removable medium 811. When the computer program is executed by central processing unit (CPU) 801, it performs the functions defined above in the system of this invention.

[0139] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0140] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0141] The modules described in the embodiments of the present invention can be implemented in software or hardware. The described modules can also be housed in a processor; for example, a processor may be described as including an interaction module and a prediction processing module. The names of these modules do not necessarily limit the module itself; for example, the interaction module may also be described as "a module that receives distribution prediction requests for logistics routes."

[0142] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs that, when executed by the device, cause the device to include: receiving a delivery forecast request for a logistics route, wherein the delivery forecast request includes a route identifier and a forecast time period for the logistics route; acquiring dynamic feature data and static feature data related to the route identifier and the forecast time period, wherein the dynamic feature data includes order features, warehouse production features, and order sorting features related to online real-time item orders; and using a pre-trained prediction model, the dynamic feature data, and the static feature data, predicting logistics delivery information corresponding to the indicated departure times and departure frequencies for the forecast time period.

[0143] According to the technical solution of the present invention, after receiving a delivery prediction request including the route identifier and predicted time period of the logistics route, the system uses a pre-trained prediction model and acquired dynamic and static feature data related to the route identifier and predicted time period, including order features, warehouse production features, and order sorting features related to online real-time item orders, to predict the logistics delivery information corresponding to the indicated departure schedule and departure frequency for the predicted time period. Since the prediction process incorporates dynamic feature data including order features, warehouse production features, and order sorting features related to online real-time item orders, and since logistics delivery is mainly for items sold on online e-commerce platforms, the solution provided in this application can effectively improve the flexibility of logistics delivery management and increase logistics delivery efficiency by associating departure schedules and departure frequencies with orders.

[0144] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A logistics distribution management method, characterized in that, include: Receive a delivery forecast request for a logistics route, wherein the delivery forecast request includes the route identifier of the logistics route and the forecast time period; Acquire dynamic and static feature data related to the route identifier and the predicted time period, wherein the dynamic feature data includes order features, warehouse production features, and order sorting features related to online real-time item orders; Using a pre-trained prediction model, the dynamic feature data, and the static feature data, logistics distribution information corresponding to the indicated departure times and departure frequencies for the predicted time period is predicted.

2. The logistics distribution management method according to claim 1, characterized in that, The acquisition of dynamic feature data related to the line identifier and the predicted time period includes: Based on the sending and receiving addresses of online real-time item orders generated within a set time period, target online real-time item orders that match the route identifier are filtered out, and based on the item information included in the target online real-time item orders, the package quantity, package volume, and logistics waybill volume are predicted. Obtain real-time production data of the warehouse used to produce the online real-time item orders, and extract warehouse feature values ​​corresponding to the warehouse production features from the real-time production data; Obtain order sorting data for the sorting process used to process the online real-time item orders, and extract the sorting feature values ​​corresponding to the order sorting features from the order sorting data.

3. The logistics distribution management method according to claim 1, characterized in that, Obtain static feature data related to the line identifier and the predicted time period, including: Based on the predicted time period, determine the historical time period; Based on the logistics and distribution data corresponding to the historical time period, analyze the route characteristics and route profiles corresponding to the route identifier.

4. The logistics distribution management method according to claim 1, characterized in that, Also includes: Based on the route, online historical item orders are grouped according to historical package volume, historical package size, and historical waybill volume; According to the route, extract the warehouse historical feature value corresponding to the warehouse production feature from the warehouse historical production data, and extract the sorting historical feature value corresponding to the order sorting feature from the order sorting historical data; Using the grouping results, the warehouse's historical feature values, the sorting's historical feature values, and the route information, the model containing the decision tree is iteratively trained to obtain the prediction model.

5. The logistics distribution management method according to claim 4, characterized in that, Also includes: For each iteration of the iterative training of the model containing the decision tree, perform the following operation: Adjust the warehouse production characteristics and / or the order sorting characteristics, and use the grouping results and the warehouse historical feature values ​​corresponding to the adjusted warehouse production characteristics and / or the sorting historical feature values ​​corresponding to the adjusted order sorting characteristics to train a model containing decision trees; Analyze the correlation between the adjusted warehouse production characteristics and the adjusted order sorting characteristics and logistics distribution management; Based on the analysis results, the warehouse production characteristics and / or order sorting characteristics used to train the model are revised.

6. The logistics distribution management method according to claim 3, characterized in that, The line characteristics include any one or more of the following characteristics: Data distribution of route type, relationships between routes, historical transport volume, historical train number, and historical load factor; And / or, The route profile includes the operational status and profit and loss information corresponding to the route identifier.

7. The logistics distribution management method according to claim 1, characterized in that, The prediction corresponds to the logistics distribution information of the indicated departure times and frequencies for the predicted time period, including: The pre-trained prediction model uses the dynamic feature data and the static feature data, including the warehouse production features and the order sorting features, to analyze multiple departure shifts, multiple departure frequencies and vehicle loads, and predicts the probability corresponding to each departure shift, departure frequency and vehicle load.

8. The logistics distribution management method according to claim 7, characterized in that, The prediction corresponds to the logistics distribution information of the indicated departure times and frequencies for the predicted time period, and also includes: Filter and output the target departure times, target departure frequencies, and target vehicle loads corresponding to predicted probabilities that are not less than a preset threshold.

9. A logistics distribution management device, characterized in that, include: The interaction module and the prediction processing module, among which, The interaction module is used to receive a delivery forecast request for a logistics route, wherein the delivery forecast request includes the route identifier of the logistics route and the forecast time period. The prediction processing module is used to acquire dynamic feature data and static feature data related to the route identifier and the prediction time period. The dynamic feature data includes order features, warehouse production features and order sorting features related to online real-time item orders. Using a pre-trained prediction model, the dynamic feature data and the static feature data, the module predicts the logistics distribution information corresponding to the indicated departure times and departure frequencies for the prediction time period.

10. A logistics distribution management system, characterized in that, include: The online order management terminal and the logistics distribution management device as described in claim 9, wherein, The online order management terminal is used to provide the logistics distribution management device with dynamic data related to orders; The logistics distribution management device is further used to acquire dynamic feature data related to route identification and predicted time period based on dynamic data related to orders, wherein the dynamic feature data includes order features, warehouse production features and order sorting features related to online real-time item orders.

11. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-8.

12. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-8.

Images