CN-121981347-A - Big data-based distribution task processing method

CN121981347ACN 121981347 ACN121981347 ACN 121981347ACN-121981347-A

Abstract

The invention relates to the technical field of distribution task processing, in particular to a distribution task processing method based on big data, which comprises the steps of responding to the fact that the number of current distribution tasks and expected distribution time length of an object to be distributed meet preset standards, determining a current distribution area of the object to be distributed based on the current distribution tasks, determining a plurality of alternative distribution tasks based on the current distribution area, determining a plurality of task clustering groups, determining distribution complexity corresponding to each task clustering group, determining a key task clustering group, determining a plurality of candidate action routes based on each key distribution task and the current distribution task, determining route evaluation values corresponding to each candidate action route based on the distribution condition of action nodes of each candidate action route and the historical distribution tasks of the object to be distributed, and determining the distribution priority of each key distribution task based on the key action route. The invention can improve the efficiency of completing the distribution task.

Inventors

WANG LI
WANG KUN
WANG MENGQI

Assignees

北京衡创智成自动化科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251205

Claims (10)

1. The big data-based distribution task processing method is characterized by comprising the following steps of: Determining a current distribution area of the object to be distributed based on the current distribution task in response to the fact that the number of the current distribution tasks of the object to be distributed and the expected distribution duration meet preset standards; determining a plurality of alternative distribution tasks based on the current distribution area, and performing cluster analysis based on each alternative distribution task to determine a plurality of task cluster groups, wherein each cluster group comprises a plurality of alternative distribution tasks; Determining distribution complexity corresponding to each task cluster group based on task information of each candidate distribution task in each task cluster group so as to determine a key task cluster group, wherein the key task cluster group comprises a plurality of key distribution tasks, and the task information comprises a distribution starting point, a distribution terminal point, a recommended distribution time window and distribution article characteristics; determining a plurality of candidate action routes based on the key delivery tasks and the current delivery tasks, wherein each candidate action route comprises a plurality of action nodes, and each action node corresponds to a delivery starting point or a delivery ending point of the delivery task; determining route evaluation values corresponding to the candidate action routes based on the action node distribution conditions of the candidate action routes and the historical distribution tasks of the objects to be distributed so as to determine key action routes; And determining the distribution priority of each key distribution task based on the key action route so as to improve the distribution task completion efficiency.
2. The big data based distribution task processing method according to claim 1, wherein the preset standard is that the number of current distribution tasks is smaller than a preset number and the expected distribution duration is smaller than a preset duration.
3. The big data based distribution task processing method according to claim 2, wherein the determining the current distribution area of the object to be distributed includes: determining a minimum surrounding area corresponding to the current delivery task based on the delivery starting point and the delivery ending point of the current delivery task; and determining the current distribution area based on the minimum surrounding area and the current distribution task number.
4. A big data based distribution task processing method according to claim 3, wherein determining a number of said alternative distribution tasks comprises: Determining a plurality of candidate delivery tasks based on the delivery tasks to be distributed of the delivery starting point in the current delivery area; Determining the maximum distribution distance of the object to be distributed based on the historical distribution task of the object to be distributed; and determining a plurality of candidate distribution tasks based on the comparison result of the distribution distance of each candidate distribution task and the maximum distribution distance.
5. The big data based distribution task processing method according to claim 4, wherein determining the distribution complexity corresponding to any one of the task clusters includes: determining a distance complex characterization value corresponding to the task cluster group based on a distribution starting point and a distribution end point of each candidate distribution task in the task cluster group; Determining a complex characterization value of the time length corresponding to the task cluster group based on recommended distribution time windows of the candidate distribution tasks in the task cluster group; Determining an article complex characterization value corresponding to the task cluster group based on the article distribution characteristics of each alternative distribution task in the task cluster group; And determining the distribution complexity corresponding to the task cluster group based on the distance complex characterization value, the duration complex characterization value and the article complex characterization value.
6. The big data based distribution task processing method according to claim 5, wherein determining the mission-critical cluster group includes: And determining the key task cluster group based on the comparison result of the distribution complexity corresponding to each task cluster group and the preset complexity.
7. The big data based distribution task processing method according to claim 6, wherein determining a number of the candidate action routes includes: Determining a plurality of action nodes based on the current distribution task, the distribution starting point and the distribution ending point of each key distribution task; Constructing node constraint relations of all the action nodes based on the recommended delivery time window of the current delivery task and all the key delivery tasks; Determining a plurality of recommended connection modes of each action node based on the node constraint relation of each action node so as to determine a plurality of candidate action routes.
8. The big data based distribution task processing method according to claim 7, wherein determining a route evaluation value corresponding to any one of the candidate action routes includes: Determining route topology features and route efficiency features corresponding to the candidate action routes based on the action node distribution conditions of the candidate action routes, wherein the route topology features comprise node distribution density, route detour and time window urgency, and the route efficiency features comprise route total distance and route total time consumption; Determining route matching features corresponding to the candidate action routes based on the historical delivery tasks of the objects to be distributed, wherein the route matching features comprise regional familiarity and delivery completion degree; And determining a route evaluation value corresponding to the candidate action route based on the route topology feature, the route efficiency feature and the route matching feature.
9. The big data based distribution task processing method according to claim 8, wherein determining the critical action route includes: Determining a key action route determining mode based on the comparison result of the route evaluation value corresponding to each candidate action route and the preset evaluation value, wherein the key action route determining mode comprises a first determining mode and a second determining mode; in the first determining manner, determining a candidate action route corresponding to the maximum route evaluation value in the candidate action routes as a key action route; In a second determination mode, the key distribution tasks corresponding to a plurality of action nodes at the tail end of the candidate action routes corresponding to the minimum route evaluation value are screened out, so that a plurality of candidate action routes are redetermined, and the candidate action route corresponding to the maximum route evaluation value in the candidate action routes is redetermined and is determined as the key action route.
10. The big data based distribution task processing method according to claim 9, wherein determining the distribution priority of each of the key distribution tasks includes: and determining the delivery priority of each key delivery task based on the connection sequence of each action node corresponding to each delivery starting point in the key action route.

Description

Big data-based distribution task processing method Technical Field The invention relates to the technical field of distribution task processing, in particular to a distribution task processing method based on big data. Background With deep integration of e-commerce, instant retail and entity economy, distribution business has been transformed from single scattered unidirectional batch task+dynamic scene, and especially in the scenes of e-commerce promotion, merchant batch delivery, trans-regional allocation and the like, the distribution task quantity is exponentially increased, and strict requirements are put on the dispatching efficiency, precision and resource suitability. The traditional distribution method of the distribution tasks mainly relies on simple geographical area division and manual experience, dynamic parameters such as the current order holding quantity of the distribution staff, the estimated distribution time length of the received orders and the like are not combined, the distribution tasks are distributed to the distribution staff in a batch binding mode, a large amount of backlog and overtime of the orders of the distribution staff are caused, and the distribution task completion efficiency is low. Chinese patent application publication No. CN105719009a discloses a method and apparatus for processing a delivery task. According to the embodiment of the invention, the M delivery tasks to be distributed are obtained, and the similarity of every two delivery tasks in the M delivery tasks is obtained according to the M delivery tasks, so that the M delivery tasks can be subjected to grouping processing according to the similarity of every two delivery tasks to obtain N delivery task groups, and each delivery task group in the N delivery task groups is distributed to delivery personnel for delivery by the delivery personnel. The prior art has the following problems that only two groups of tasks are distributed for distribution staff to distribute, only the similarity of the distribution task groups is considered, the completion degree of the whole distribution task is not considered, and the distribution task completion efficiency is low. Disclosure of Invention Therefore, the invention provides a distribution task processing method based on big data, which is used for solving the problem of lower distribution task completion efficiency in the prior art. In order to achieve the above object, the present invention provides a big data based distribution task processing method, including: Determining a current distribution area of the object to be distributed based on the current distribution task in response to the fact that the number of the current distribution tasks of the object to be distributed and the expected distribution duration meet preset standards; determining a plurality of alternative distribution tasks based on the current distribution area, and performing cluster analysis based on each alternative distribution task to determine a plurality of task cluster groups, wherein each cluster group comprises a plurality of alternative distribution tasks; Determining distribution complexity corresponding to each task cluster group based on task information of each candidate distribution task in each task cluster group so as to determine a key task cluster group, wherein the key task cluster group comprises a plurality of key distribution tasks, and the task information comprises a distribution starting point, a distribution terminal point, a recommended distribution time window and distribution article characteristics; determining a plurality of candidate action routes based on the key delivery tasks and the current delivery tasks, wherein each candidate action route comprises a plurality of action nodes, and each action node corresponds to a delivery starting point or a delivery ending point of the delivery task; determining route evaluation values corresponding to the candidate action routes based on the action node distribution conditions of the candidate action routes and the historical distribution tasks of the objects to be distributed so as to determine key action routes; And determining the distribution priority of each key distribution task based on the key action route so as to improve the distribution task completion efficiency. Further, the preset standard is that the number of the current delivery tasks is smaller than the preset number and the expected delivery duration is smaller than the preset duration. Further, the determining the current distribution area of the object to be distributed includes: determining a minimum surrounding area corresponding to the current delivery task based on the delivery starting point and the delivery ending point of the current delivery task; and determining the current distribution area based on the minimum surrounding area and the current distribution task number. Further, determining a number of the alternative dispensing tasks include