CN113537853A - Order distribution method, order distribution device, readable storage medium and electronic equipment - Google Patents

Order distribution method, order distribution device, readable storage medium and electronic equipment Download PDF

Info

Publication number
CN113537853A
CN113537853A CN202010292165.6A CN202010292165A CN113537853A CN 113537853 A CN113537853 A CN 113537853A CN 202010292165 A CN202010292165 A CN 202010292165A CN 113537853 A CN113537853 A CN 113537853A
Authority
CN
China
Prior art keywords
characteristic information
path
distribution
planned path
delivery
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010292165.6A
Other languages
Chinese (zh)
Inventor
王莉
包建东
王圣尧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Sankuai Online Technology Co Ltd
Original Assignee
Beijing Sankuai Online Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Sankuai Online Technology Co Ltd filed Critical Beijing Sankuai Online Technology Co Ltd
Priority to CN202010292165.6A priority Critical patent/CN113537853A/en
Publication of CN113537853A publication Critical patent/CN113537853A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0835—Relationships between shipper or supplier and carriers
    • G06Q10/08355—Routing methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
    • G06Q30/0635—Processing of requisition or of purchase orders

Landscapes

  • Business, Economics & Management (AREA)
  • Economics (AREA)
  • Engineering & Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • Accounting & Taxation (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Human Resources & Organizations (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The disclosure relates to an order distribution method, an order distribution device, a readable storage medium and an electronic device. The method comprises the following steps: determining a plurality of delivery capacities associated with the orders to be distributed; respectively planning a task point path for each distribution capacity according to the order to be distributed and the current distribution task of each distribution capacity to obtain a respective planned path of each distribution capacity; for each planned path, determining robustness characteristic information corresponding to the planned path according to the probability distribution of the random distribution characteristic information corresponding to the planned path; determining an optimal planning path according to the robustness characteristic information; and distributing the orders to be distributed to the distribution capacity corresponding to the optimal planning path. The probability distribution of the random distribution characteristic information corresponding to the planning path is considered when the order is distributed, so that the accuracy of order distribution can be improved. And the orders to be distributed are distributed to the distribution capacity corresponding to the optimal planning path, so that the distribution efficiency of the distribution capacity is improved.

Description

Order distribution method, order distribution device, readable storage medium and electronic equipment
Technical Field
The present disclosure relates to the field of network technologies, and in particular, to an order allocation method, an order allocation apparatus, a readable storage medium, and an electronic device.
Background
With the rapid development of network technology, network applications such as express delivery, takeaway delivery, shared charge pal delivery and the like are widely popularized, and users can place orders through the network and complete delivery services through delivery capacity. In order to improve the delivery efficiency, the order scheduling system usually makes an order dispatching decision according to the matching degree of the order to be distributed and the delivery capacity.
In general, when analyzing the matching degree of the order to be distributed and the delivery capacity, deterministic information and stochastic information related to delivery need to be considered. Deterministic information may include merchant location, user location, order price, projected time of delivery, etc., and stochastic information may include merchant meal time, travel time, meal-pick-up completion time, and delivery completion time. The delivery completion time may be the delivery capacity or the user, and therefore, the delivery time may be the delivery time related to the delivery capacity behavior or the delivery time related to the user behavior. Because the influence factors of the random information are numerous, the real data are difficult to obtain, and even the machine learning means is difficult to accurately estimate. Therefore, in the conventional order scheduling system, in order to simplify the problem, randomness information is generally assumed as deterministic data to be processed so as to analyze the matching degree between the order to be distributed and the delivery capacity, so that the matching degree between the order to be distributed and the delivery capacity cannot be accurately determined, an accurate order dispatching decision cannot be made, and the delivery efficiency of the delivery capacity is affected.
Disclosure of Invention
The purpose of the present disclosure is to provide an order distribution method, an order distribution device, a readable storage medium, and an electronic device, so as to improve the distribution efficiency of the distribution capacity.
In order to achieve the above object, the present disclosure provides an order allocation method, including:
determining a plurality of delivery capacities associated with the orders to be distributed;
respectively planning a task point path for each distribution capacity according to the order to be distributed and the current distribution task of each distribution capacity to obtain a respective planned path of each distribution capacity;
for each planned path, determining robustness characteristic information corresponding to the planned path according to the probability distribution of the random distribution characteristic information corresponding to the planned path, wherein the robustness characteristic information is used for representing the influence degree of the order to be distributed on the delivery time of the distribution capacity corresponding to the planned path;
determining an optimal planning path according to the robustness characteristic information;
and distributing the order to be distributed to the distribution capacity corresponding to the optimal planning path.
Optionally, the random distribution characteristic information includes at least one of: article preparation completion time; the fetching completion time; the travel time of the delivery capacity; item delivery time.
Optionally, the determining, according to the probability distribution of the random distribution characteristic information corresponding to the planned path, the robustness characteristic information corresponding to the planned path includes:
acquiring path characteristic information of the planned path, wherein the path characteristic information of the planned path comprises distribution parameters of probability distribution of random distribution characteristic information corresponding to the planned path;
inputting the path characteristic information of the planned path into a robustness characteristic information determination model to obtain the robustness characteristic information corresponding to the planned path.
Optionally, the path feature information further includes: the number of task points in the path and/or sequence indication information for indicating the sequence of the task points in the path.
Optionally, the robust feature information determination model is obtained by:
acquiring path characteristic information of a historical path, wherein the path characteristic information of the historical path comprises a distribution parameter of probability distribution of random distribution characteristic information corresponding to the historical path;
acquiring robustness characteristic information corresponding to the historical path;
and training a neural network model by taking the path characteristic information of the historical path as a model input parameter and taking the robustness characteristic information corresponding to the historical path as a model output parameter so as to obtain the robustness characteristic information determination model.
Optionally, the obtaining robustness characteristic information corresponding to the historical path includes:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information included in the path characteristic information of the historical path, wherein different sampling values of the random distribution characteristic information are used for forming different simulation scenes of the historical path;
determining delivery time characteristic information of the historical path in each simulation scene;
and determining robustness characteristic information corresponding to the historical path according to the delivery time characteristic information of the historical path in each simulated scene.
Optionally, the determining the delivery time characteristic information of the historical path in each simulation scenario includes:
determining whether overtime orders exist in the orders related to the historical path under each simulation scene according to the estimated delivery time of each order related to the historical path under each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the historical path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the historical path, and taking the average distribution margin as the delivery time characteristic information of the historical path in the simulation scene.
Optionally, before the step of obtaining the path feature information of the planned path, the method further includes:
and determining that the random distribution characteristic information corresponding to the planned path does not belong to preset abnormal information, and determining that the environment information corresponding to the planned path does not belong to preset abnormal environment information.
Optionally, the determining, according to the probability distribution of the random distribution characteristic information corresponding to the planned path, the robustness characteristic information corresponding to the planned path includes:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path, wherein different sampling values of the random distribution characteristic information are used for forming different simulation scenes of the planned path;
determining delivery time characteristic information of the planned path in each simulation scene;
and determining robustness characteristic information corresponding to the planned path according to the delivery time characteristic information of the planned path in each simulated scene.
Optionally, the determining delivery time characteristic information of the planned path in each simulation scenario includes:
determining whether an overtime order exists in the orders related to the planned path in each simulation scene according to the estimated delivery time of each order related to the planned path in each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the planned path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the planned path, and taking the average distribution margin as the delivery time characteristic information of the planned path in the simulation scene.
Optionally, the sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path includes:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path and a target sampling number, wherein the target sampling number is an optimal sampling number determined from a plurality of preset sampling numbers according to the probability distribution of the random distribution characteristic information corresponding to a historical path and robustness characteristic information corresponding to the historical path under each preset sampling number.
Optionally, before the step of sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path, the method further includes:
and determining that the random distribution characteristic information corresponding to the planned path belongs to preset abnormal information, or determining that the environment information corresponding to the planned path belongs to preset abnormal environment information.
Optionally, the robustness characteristic information includes at least one of: an expectation of the delivery time characteristic information for each of the simulated scenarios, a variance of the delivery time characteristic information for each of the simulated scenarios, and an expectation of the delivery time characteristic information for the simulated scenarios where there is a time-out order.
The second aspect of the present disclosure also provides an order distribution apparatus, including:
a first determination module configured to determine a plurality of delivery capacities associated with an order to be allocated;
the planning module is configured to perform task point path planning on each delivery capacity respectively according to the order to be distributed and the current delivery task of each delivery capacity to obtain a respective planned path of each delivery capacity;
a second determining module, configured to determine, for each planned path, robustness characteristic information corresponding to the planned path according to a probability distribution of random delivery characteristic information corresponding to the planned path, where the robustness characteristic information is used to characterize a degree of influence of the order to be allocated on delivery time of delivery capacity corresponding to the planned path;
a third determining module configured to determine an optimal planned path according to the robustness characteristic information;
and the distribution module is configured to distribute the orders to be distributed to the distribution capacity corresponding to the optimal planning path.
Optionally, the random distribution characteristic information includes at least one of: article preparation completion time; the fetching completion time; the travel time of the delivery capacity; item delivery time.
Optionally, the second determining module includes:
a first obtaining sub-module, configured to obtain path feature information of the planned path, where the path feature information of the planned path includes a distribution parameter of a probability distribution of random distribution feature information corresponding to the planned path;
and the second obtaining submodule is configured to input the path characteristic information of the planned path to a robustness characteristic information determination model, and obtain the robustness characteristic information corresponding to the planned path.
Optionally, the path feature information further includes: the number of task points in the path and/or sequence indication information for indicating the sequence of the task points in the path.
Optionally, the apparatus further comprises:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is configured to be used for acquiring path characteristic information of a historical path, and the path characteristic information of the historical path comprises a distribution parameter of probability distribution of random distribution characteristic information corresponding to the historical path;
the second acquisition module is configured to acquire robustness characteristic information corresponding to the historical path;
and the training module is configured to train a neural network model by taking the path characteristic information of the historical path as a model input parameter and taking the robustness characteristic information corresponding to the historical path as a model output parameter so as to obtain the robustness characteristic information determination model.
Optionally, the second obtaining module includes:
a first sampling sub-module, configured to sample the random distribution feature information according to a probability distribution of the random distribution feature information included in the path feature information of the historical path, where different sampling values of the random distribution feature information are used to form different simulation scenarios of the historical path;
a first determining sub-module configured to determine delivery time characteristic information of the historical path in each simulation scenario;
and the second determining submodule is configured to determine robustness characteristic information corresponding to the historical path according to the delivery time characteristic information of the historical path in each simulated scene.
Optionally, the first determining submodule is configured to:
determining whether overtime orders exist in the orders related to the historical path under each simulation scene according to the estimated delivery time of each order related to the historical path under each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the historical path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the historical path, and taking the average distribution margin as the delivery time characteristic information of the historical path in the simulation scene.
Optionally, the apparatus further comprises:
the fourth determining module is configured to determine that the random distribution characteristic information corresponding to the planned path does not belong to preset abnormal information, and the environment information corresponding to the planned path does not belong to preset abnormal environment information.
Optionally, the second determining module includes:
the second sampling submodule is configured to sample the random distribution characteristic information according to probability distribution of the random distribution characteristic information corresponding to the planned path, wherein different sampling values of the random distribution characteristic information are used for forming different simulation scenes of the planned path;
a third determining submodule configured to determine delivery time characteristic information of the planned path in each simulation scenario;
and the fourth determining submodule is configured to determine robustness characteristic information corresponding to the planned path according to the delivery time characteristic information of the planned path in each simulated scene.
Optionally, the third determining submodule is configured to:
determining whether an overtime order exists in the orders related to the planned path in each simulation scene according to the estimated delivery time of each order related to the planned path in each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the planned path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the planned path, and taking the average distribution margin as the delivery time characteristic information of the planned path in the simulation scene.
Optionally, the second sampling sub-module is configured to:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path and a target sampling number, wherein the target sampling number is an optimal sampling number determined from a plurality of preset sampling numbers according to the probability distribution of the random distribution characteristic information corresponding to a historical path and robustness characteristic information corresponding to the historical path under each preset sampling number.
Optionally, the apparatus further comprises:
a fifth determining module, configured to determine that the random distribution feature information corresponding to the planned path belongs to preset abnormal information, or determine that the environment information corresponding to the planned path belongs to preset abnormal environment information.
Optionally, the robustness characteristic information includes at least one of: an expectation of the delivery time characteristic information for each of the simulated scenarios, a variance of the delivery time characteristic information for each of the simulated scenarios, and an expectation of the delivery time characteristic information for the simulated scenarios where there is a time-out order.
The third aspect of the present disclosure also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method provided by the first aspect of the present disclosure.
The fourth aspect of the present disclosure also provides an electronic device, including:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the method provided by the first aspect of the present disclosure.
According to the technical scheme, the robustness characteristic information corresponding to the planned path is determined according to the probability distribution of the random distribution characteristic information corresponding to the planned path, the optimal planned path is determined according to the robustness characteristic information, and the order to be distributed can be distributed to the distribution capacity corresponding to the optimal planned path. The probability distribution of the random distribution characteristic information corresponding to the planned path is considered when the order is distributed, so that the robustness characteristic information corresponding to the planned path can be accurately determined, the order is distributed according to the robustness characteristic information, and the accuracy of order distribution is improved. And the orders to be distributed are distributed to the distribution capacity corresponding to the optimal planning path, so that the distribution efficiency of the distribution capacity is improved.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a method of order distribution according to an exemplary embodiment.
Fig. 2 is a schematic diagram of a planned path, shown in accordance with an exemplary embodiment.
FIG. 3 is a flow chart illustrating a method of order distribution according to another exemplary embodiment.
FIG. 4 is a flow diagram illustrating a method of training a robust feature information determination model in accordance with an exemplary embodiment.
Fig. 5 is a flowchart illustrating a method for obtaining robust feature information corresponding to a historical path according to an exemplary embodiment.
FIG. 6 is a schematic diagram illustrating a method of determining the delivery margin for an order according to an exemplary embodiment.
FIG. 7 is a flow chart illustrating a method of order distribution according to another exemplary embodiment.
FIG. 8 is a block diagram illustrating an order distribution apparatus according to an exemplary embodiment.
FIG. 9 is a block diagram illustrating an electronic device in accordance with an example embodiment.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
The method is mainly applied to business scenes such as take-out delivery, supermarket delivery, same-city delivery, express delivery and the like. In these business scenarios, it is usually necessary to combine many factors to perform allocation scheduling on the to-be-allocated orders so as to allocate the to-be-allocated orders to the appropriate delivery capacity. In the related art, when an order to be allocated is allocated, randomness information is generally assumed as deterministic data to be processed so as to analyze the matching degree of the order to be allocated and the delivery capacity, and the processing mode is simple. However, the robustness of determining the planned path according to the method is difficult to guarantee. Specifically, in a real scene, the influence factor of the random information is large, the emergency situation is large, and the random information cannot be replaced by a certain scene. For example, taking the meal time as a definite value, if the merchant clicks, it will cause the dispenser to wait at the merchant. Under the condition of less delivery orders, the merchant card meal does not have worse influence. However, in a situation of a mid-day peak, which is a distribution capacity shortage, a merchant may have a card meal to cause the following problems: 1. the merchant card meal causes overtime of the order of the delivery capacity, reduces the delivery punctuality rate and influences the user experience. 2. Merchant card meals are a waste of delivery capacity. The meal delivery time of the delivery capacity in the peak noon period (for example, 10:00 to 14:00) is constant, and the longer the delivery capacity is in the time of a merchant for waiting for meals, the shorter the effective meal delivery time is, so that the delivery capacity can complete the reduction of single amount in unit time, and the delivery efficiency is seriously influenced. 3. The contradiction between the distribution capacity and the merchants is excited.
In view of this, the present disclosure provides an order distribution method, an order distribution apparatus, a readable storage medium and an electronic device to improve distribution efficiency of distribution capacity.
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
FIG. 1 is a flow chart illustrating a method of order distribution according to an exemplary embodiment. As shown in fig. 1, the method may include steps 101 to 105.
In step 101, a plurality of delivery capacities associated with an order to be allocated is determined.
In the present disclosure, the order allocation method may be executed by a server of an order scheduling system or a delivery platform, and since it needs to determine which delivery capacity to allocate an order to be allocated to from a plurality of delivery capacities, it is first necessary to determine a plurality of delivery capacities associated with the order to be allocated. Wherein the delivery capacity may be a delivery person or a delivery device (e.g., a delivery robot, an unmanned delivery vehicle, an unmanned aerial vehicle, etc.). The plurality of delivery capacities associated with the order to be distributed may be delivery capacities corresponding to a delivery location or a pickup location of the order to be distributed. For example, assuming that the delivery site of the order to be distributed is the sea area of beijing, the plurality of delivery capacities may be delivery capacities belonging to the sea area. Further, the plurality of delivery capacities associated with the to-be-allocated order may also be candidate delivery capacities preliminarily determined by the server. For example, the server may determine a matching degree of the to-be-allocated order and the delivery capacity using a correlation technique, determine the delivery capacity with the matching degree greater than a preset matching degree as a candidate delivery capacity, and determine the candidate delivery capacity as a plurality of delivery capacities associated with the to-be-allocated order.
In step 102, a task point path planning is performed on each delivery capacity according to the order to be allocated and the current delivery task of each delivery capacity, so as to obtain a respective planned path of each delivery capacity.
After determining a plurality of delivery capacities associated with the orders to be distributed, the server determines a current delivery task of the delivery capacities for each delivery capacity, wherein the current delivery task refers to a task (including an fetching task and a delivery task) corresponding to the order which is already distributed to the delivery capacities but is not completed. Since the current delivery task of the delivery capacity is the task that the delivery capacity needs to perform to complete, it must be considered when determining the planned path of the delivery capacity.
And then, determining a planned path of the delivery capacity according to the current delivery task and the to-be-distributed orders of the delivery capacity (namely, the fetching task and the delivery task of the to-be-distributed orders) and a preset path optimization algorithm. The planning path comprises a goods taking task point and a goods delivering task point. For example, assuming that the order to be allocated is C and the allocated orders are a and B, after path planning, a planned path is obtained as shown in fig. 2. In fig. 2, the triangles represent pick task points, the circles represent delivery task points, and the arrows represent the order of pick and delivery. Note that in fig. 2, the rider is currently in the process of completing the pick task for order a and completing the pick task for order B. In this manner, the server may determine a respective planned path for each delivery capacity.
In step 103, for each planned path, the robustness characteristic information corresponding to the planned path is determined according to the probability distribution of the random distribution characteristic information corresponding to the planned path.
In order to ensure that when an order to be allocated is allocated to a certain delivery capacity, the efficiency of the delivery capacity for completing the current delivery task is not affected, therefore, in the present disclosure, it may be assumed that the order to be allocated is allocated to the plurality of delivery capacities, and robustness characteristic information corresponding to a planned path of each delivery capacity is obtained, where the robustness characteristic information may be used to characterize a degree of influence of the order to be allocated on a delivery time of the delivery capacity corresponding to the planned path.
In addition, in order to solve the problem in the related art that the influence of orders to be distributed on the delivery time of the delivery capacity corresponding to the planned path cannot be accurately determined due to the fact that the randomness information is assumed to be processed as the deterministic information, the robustness characteristic information corresponding to the planned path is determined according to the probability distribution of the random delivery characteristic information corresponding to the planned path. The present disclosure does not specifically limit the manner of determining the probability distribution of the randomly distributed feature information.
In the present disclosure, the random distribution characteristic information may include at least one of: article preparation completion time; the fetching completion time; the travel time of the delivery capacity; item delivery time. The item preparation completion time refers to the time from the time when the merchant receives the order to the time when the merchant prepares the item; the fetching completion time refers to the time length from the time of arrival of the distribution capacity at the merchant to the time of fetching the goods; the moving time of the distribution capacity refers to the time length of the distribution capacity from one task point to another task point; the delivery time of an item refers to the length of time that the delivery capacity takes from reaching the appointed delivery location to delivering the item to the user. For example, assuming the order is a take out order, the probability distribution of item preparation completion times may be a probability distribution of merchant meal times, where the probability distribution may be determined based on the merchant's historical meal times.
It should be noted that, when the random distribution characteristic information includes the time for completing preparation of the article, the probability distribution of the random distribution characteristic information corresponding to the planned path may be the probability distribution of the meal time of the merchant of each merchant included in the planned path, or may be the probability distribution of the meal time of the merchant of one or several merchants among multiple merchants included in the planned path, which is not specifically limited in this disclosure.
In step 104, an optimal planned path is determined according to the robustness characteristic information.
In step 105, the order to be allocated is allocated to the delivery capacity corresponding to the optimal planned path.
As described above, the robustness characteristic information may be used to characterize the degree of influence of the to-be-allocated order on the delivery time of the delivery capacity corresponding to the planned path, and therefore, the planned path with the smallest influence of the to-be-allocated order on the delivery time of the delivery capacity, that is, the planned path with the smallest influence on the on-time delivery of the delivery capacity, may be determined according to the robustness characteristic information, and the planned path is determined as the optimal planned path, so that the to-be-allocated order is allocated to the delivery capacity corresponding to the optimal planned path.
According to the technical scheme, the robustness characteristic information corresponding to the planned path is determined according to the probability distribution of the random distribution characteristic information corresponding to the planned path, the optimal planned path is determined according to the robustness characteristic information, and the order to be distributed can be distributed to the distribution capacity corresponding to the optimal planned path. The probability distribution of the random distribution characteristic information corresponding to the planned path is considered when the order is distributed, so that the robustness characteristic information corresponding to the planned path can be accurately determined, the order is distributed according to the robustness characteristic information, and the accuracy of order distribution is improved. And the orders to be distributed are distributed to the distribution capacity corresponding to the optimal planning path, so that the distribution efficiency of the distribution capacity is improved.
A specific manner of determining the robustness characteristic information corresponding to the planned path according to the probability distribution of the random distribution characteristic information corresponding to the planned path in step 103 will be described below.
First, a detailed description will be given of a specific embodiment of determining robust feature information corresponding to a planned path by using a machine learning method. Specifically, as shown in fig. 3, the step 103 may include steps 1031 and 1032.
In step 1031, path feature information of the planned path is acquired. The path characteristic information of the planned path comprises a distribution parameter of probability distribution of random distribution characteristic information corresponding to the planned path.
The probability distribution of the randomly distributed feature information may be a continuous probability distribution (e.g., a normal distribution) or a discrete probability distribution (e.g., a binomial distribution, a poisson distribution, etc.). In order to input the path feature information of the planned path into the model, in the present disclosure, the path feature information of the planned path includes a distribution parameter of a probability distribution of random distribution feature information corresponding to the planned path. For example, when the probability distribution of the random distribution characteristic information is normal distribution, the distribution parameters of the probability distribution of the random distribution characteristic information are the mean value of the random distribution characteristic information and the variance of the random distribution characteristic information; under the condition that the probability distribution of the random distribution characteristic information is binomial distribution, the distribution parameters of the probability distribution of the random distribution characteristic information are the number of terms and the probability of the random distribution characteristic information.
The path characteristic information of the planned path may include, in addition to a distribution parameter of a probability distribution of the random distribution characteristic information corresponding to the planned path, the number of task points in the path and/or sequence indication information for indicating a sequence of the task points in the path. In addition, in other embodiments, the path characteristic information of the planned path may further include other information, which is not specifically limited by the present disclosure.
In step 1032, the path feature information of the planned path is input to the robustness feature information determination model, and robustness feature information corresponding to the planned path is obtained.
In the machine learning mode, the path characteristic information of the planned path is input to the robustness characteristic information determination model, and the robustness characteristic information corresponding to the planned path output by the robustness characteristic information determination model can be obtained.
By adopting the technical scheme, the path characteristic information of the planned path is input into the robustness characteristic information determination model in a machine learning mode, so that the robustness characteristic information corresponding to the planned path output by the robustness characteristic information determination model can be obtained, the process of determining the robustness characteristic information corresponding to the planned path is simplified, and the convenience of determining the robustness characteristic information corresponding to the planned path is improved.
In addition, when the robustness characteristic information determination model is trained, the training samples are mostly normal samples, wherein the normal samples refer to historical paths in which the merchant cannot have long calorie meal time or the environment has little influence on the moving time or delivery time of the distribution capacity. When the planned path is normal, the robustness characteristic information corresponding to the planned path can be accurately determined according to the robustness characteristic information determination model, and when the planned path is abnormal, the robustness characteristic information corresponding to the planned path cannot be accurately determined through the robustness characteristic information determination model. Therefore, in a preferred embodiment, before the step of obtaining the path characteristic information of the planned path, it is determined that the random distribution characteristic information corresponding to the planned path does not belong to the preset abnormal information, and the environment information corresponding to the planned path does not belong to the preset abnormal environment information. The preset abnormal information may be that the random distribution characteristic information is greater than a preset value, for example, the time of a merchant meal is greater than a time threshold, and the like. The preset abnormal environment may be bad weather such as snowstorm weather, rainstorm weather, and the like.
By adopting the scheme, the robustness characteristic information corresponding to the planned path is determined in a machine learning mode only under the condition that the random distribution characteristic information corresponding to the planned path does not belong to the preset abnormal information and the environment information corresponding to the planned path does not belong to the preset abnormal environment information, so that the accuracy of determining the robustness characteristic information corresponding to the planned path is further improved.
The following describes a training method of the robustness characteristic information determination model used in the above. As shown in fig. 4, the training method may include steps 401 to 403.
In step 401, path feature information of a historical path is acquired.
In the process of training the robustness characteristic information determination model, firstly, path characteristic information of a historical path is obtained, wherein the path characteristic information of the historical path comprises distribution parameters of probability distribution of random distribution characteristic information corresponding to the historical path. Wherein the historical path may be a planned path of delivery capacity over a past time (e.g., a past year, half year, month, etc.).
In step 402, robustness feature information corresponding to the historical path is obtained.
As shown in fig. 5, a specific embodiment of acquiring robustness characteristic information corresponding to a historical path may include steps 4021 to 4023.
In step 4021, sampling the random distribution characteristic information according to a probability distribution of the random distribution characteristic information included in the path characteristic information of the historical path, where different sampling values of the random distribution characteristic information are used to form different simulation scenes of the historical path.
It should be noted that, for a historical path, since each task point on the path has been completed, the corresponding random distribution feature information is a fixed numerical value. If the robustness characteristic information corresponding to the historical path is determined only according to the random distribution characteristic information of the historical path, the random distribution characteristic information is treated as a fixed numerical value to obtain the robustness characteristic information corresponding to the historical path. As described in the background, with this method, the accuracy of the determined robust feature information cannot be guaranteed. Therefore, in the present disclosure, in order to embody the randomness of the random distribution characteristic information, the server may sample the random distribution characteristic information according to the probability distribution of the random distribution characteristic information to randomly obtain a plurality of sampling values, and form different simulation scenarios of the historical path according to the plurality of sampling values.
For convenience of subsequent description, in the present disclosure, the order is taken as a takeaway order, and the random distribution characteristic information includes a meal time of the merchant as an example, which is described later.
For example, assume that the historical path includes order 1, order 2, and order 3, each order includes a meal taking task point and a meal delivery task point, and each meal taking task point corresponds to a merchant, where the probability distribution of the meal delivery time of the merchant of each merchant may be the same or different. The server can sample the restaurant food serving time of the three merchants according to the probability distribution of the restaurant food serving time of the three merchants. For example, sampling the merchant meal delivery time of the merchant 1 to obtain the merchant meal delivery time t1, sampling the merchant meal delivery time of the merchant 2 to obtain the merchant meal delivery time t2, and sampling the merchant meal delivery time of the merchant 3 to obtain the merchant meal delivery time t3, then a simulation scene on the historical road can be formed according to the sampled merchant meal delivery times t1, t2, and t 3. That is, sampling the meal delivery time of the 3 merchants once can obtain a simulation scene, and thus sampling the meal delivery time of the 3 merchants K times to obtain K simulation scenes.
It should be noted that, since the random distribution characteristic information is described as including only the meal time of the merchant in this example, other random distribution characteristic information (for example, the fetching completion time, the moving time of the distribution capacity, and the item delivery time) may be a real fixed value corresponding to the historical path. It should be understood by those skilled in the art that the random distribution characteristic information may further include multiple ones of an article preparation completion time, an article taking completion time, a moving time of the distribution capacity, and an article delivery time, and in case of multiple ones, the sampling manner is similar to that described above, and will not be described herein again.
In addition, each simulation scenario in this disclosure involves the same order completion as on the historical path, and the number and order of task points is the same. The difference is that the random delivery characteristic information corresponding to the historical path is a fixed true value (for example, the meal delivery time of each merchant on the historical path is the true meal delivery time of the merchant when the previous delivery capacity completes the task on the historical path), and the meal delivery time of each merchant in the simulation scene may be a non-true value (for example, the historical meal delivery time of a certain merchant is 4min, 5min, and 6min, respectively, and the meal delivery time of the merchant in the simulation scene may be 4.5 min). In addition, the larger the number of simulation scenes is, the more randomness of the randomly distributed characteristic information can be reflected.
In step 4022, the delivery time characteristic information of the historical path in each simulation scenario is determined.
After different simulation scenes of the historical path are obtained according to the scheme, delivery time characteristic information under each simulation scene can be respectively determined, wherein the delivery time characteristic information can be used for representing the capacity of delivering the on-time delivery order of the transport capacity under the simulation scene. The step of determining the delivery time characteristic information in each simulation scene may include the following steps:
(1): and determining whether the overtime orders exist in the orders related to the historical path in each simulation scene according to the estimated delivery time of each order related to the historical path in each simulation scene.
As described above, while obtaining the simulation scene, the relevant information of the simulation scene, such as the meal-out time of the merchant, the meal-taking completion time of the delivery capacity, the moving time of the delivery capacity, the delivery completion time, and the like, may be determined, and the estimated delivery time of each order in the simulation scene may be determined according to the relevant information of the simulation scene. Wherein the estimated lead time of each order may be determined using a related art technique, the disclosure is not particularly limited.
Thereafter, it may be determined whether there is a timed-out order in the orders involved in the historical path in each simulation scenario based on the estimated delivery time. Illustratively, continuing with the example of the historical path including order 1, order 2, and order 3, for each simulation scenario, respectively determining the estimated delivery time of order 1, order 2, and order 3 in the simulation scenario, and determining whether order 1, order 2, and order 3 are timeout orders according to the estimated delivery time and the committed delivery time (which is the delivery time fed back to the user by the server when the user places an order). Specifically, for each order, if the estimated delivery time is greater than the committed delivery time, the order is an overtime order.
(2): aiming at each simulation scene, taking the maximum overtime time as the delivery time characteristic information of the historical path in the simulation scene under the condition that an overtime order exists in the simulation scene; and in the case that no overtime orders exist in the simulation scene, determining the distribution margin of each order involved in the historical path, and taking the average distribution margin as the delivery time characteristic information of the historical path in the simulation scene.
In one embodiment, when there is a timeout order in the simulation scenario (i.e., there is an order with an estimated delivery time greater than the committed delivery time), the maximum timeout time may be used as the delivery time characteristic information of the historical path in the simulation scenario. For example, the lead time characteristic information is f (x), and the difference between the estimated lead time and the committed lead time of the order i is xiThen f (x) maxi∈IxiWherein I is the number of orders related to the historical path, and I represents the ith order.
In another embodiment, in the case where no overtime orders exist in the simulation scenario (i.e., the estimated delivery time of each order in the simulation scenario is less than or equal to the committed delivery time of the order), the delivery margin of each order involved in the historical path is determined, and the average margin is used as the delivery time characteristic information of the historical path in the simulation scenario. The delivery allowance represents the difference between the estimated delivery time and the committed delivery time.
In one embodiment, the calculated delivery allowances for each order may be averaged to obtain an average delivery allowance. However, in the delivery field, considering that orders on the same road have relevance before, for example, the delivery volume of the first order on the path is-10 (perhaps completing delivery 10min ahead), the delivery margin of the second order is-12 (perhaps completing delivery 12min ahead), and the delivery volume of the third order is-5 (perhaps completing delivery 5min ahead), wherein the shorter the time to complete delivery ahead, the greater the stress of delivery capacity. According to the empirical value, the pressure of the delivery capacity is related to the value with the minimum absolute value of the delivery allowance in a path, so that in order to better meet the practical application scene and improve the accuracy of the determined delivery time characteristic information, in another embodiment, the average delivery allowance can be determined in the following way.
Specifically, according to the delivery allowance of each order, determining an order j which is positioned behind an order i and has the smallest absolute value of the delivery allowance, and determining the delivery allowance of the order j as a target delivery allowance, wherein the initial value of i is 0, and the target delivery allowance is used as the delivery allowance of other orders (including the order j but not including the order i) positioned between the order i and the order j. And then, when the I is equal to the j, determining the order j which is positioned after the order I and has the smallest absolute value of the distribution margin, determining the distribution margin of the order j as a target distribution margin, and using the target distribution margin as the distribution margin of other orders positioned between the order I and the order j until the j is equal to the I, wherein the I is the number of orders related to the historical path. Finally, the average value of the delivery allowances of each order determined in the above manner is used as the average delivery allowances.
For example, as shown in FIG. 6, assume that the historical path involves 7 orders, and that in a simulation scenario, the delivery margin (the value below each order) for each order is as shown in FIG. 6. Referring to fig. 6, when i is 0, the absolute value of the delivery margin of order 3 is the smallest, so that the delivery margin-5 of order 3 is the target delivery margin, and the delivery margins of orders 1 to 3 are all determined to be-5. Next, when i is 3, order j, which is located after order 3 and has the smallest absolute value of the delivery allowance, is determined as order 6, and delivery allowance-7 of order 6 is determined as the delivery allowances of orders 4 to 6. Likewise, order 7 has a delivery margin of-9. Then, determining the average distribution margin under the simulation scene according to the following formula, and taking the average distribution margin as the lead time characteristic information of the historical path under the simulation scene:
f(x)＝(-5-5-5-7-7-9)7
in the above manner, for each simulated scene of the historical path, the delivery time characteristic information of the simulated scene can be determined. For example, if the historical path has 1000 simulated scenes, 1000 delivery time characteristic information may be determined for the historical path.
In step 4023, robust feature information corresponding to the historical path is determined according to the delivery time feature information of the historical path in each simulation scenario.
In the present disclosure, the robustness feature information may include at least one of: the expectation of the delivery time characteristic information in each simulation scenario, the variance of the delivery time characteristic information in each simulation scenario, and the expectation of the delivery time characteristic information in a simulation scenario in which a timeout order exists.
As described above, in step 4022, the delivery time feature information of the given historical route in different simulation scenes may be determined, and the expectation E (f (x)) of the delivery time feature information in each simulation scene, the variance D (f (x)) of the delivery time feature information in each simulation scene, and the expectation E (f) of the delivery time feature information in the simulation scene with the overtime order may be calculated from the delivery time feature information in the different simulation scenesa(x) Wherein f) isa(x) The delivery time characteristic information in the simulation scene with the overtime order exists.
It should be noted that, the robustness characteristic information may include: when one of the expectation of the delivery time characteristic information in each simulation scenario and the variance of the delivery time characteristic information in each simulation scenario is used, in step 104 in fig. 1, the planned path in which the expectation of the delivery time characteristic information in each simulation scenario is negative and the absolute value of the negative is the largest, or the variance of the delivery time characteristic information in each simulation scenario is the smallest may be determined as the optimal planned path. When the robustness characteristic information includes one of expectations of delivery time characteristic information in a simulation scenario in which a timeout order exists, the planned path in which the expectation of the delivery time characteristic information in the simulation scenario in which the timeout order exists is the minimum may be determined as the optimal planned path. When the robustness characteristic information includes multiple ones of the expectation of the delivery time characteristic information in each simulation scenario, the variance of the delivery time characteristic information in each simulation scenario, and the expectation of the delivery time characteristic information in the simulation scenario in which the timeout order exists, the multiple ones may be weighted to calculate a result, and an optimal planned path may be determined according to the result, and the like, which is not specifically limited by the present disclosure.
Returning to fig. 4, in step 403, the neural network model is trained by using the path feature information of the historical path as a model input parameter and the robustness feature information corresponding to the historical path as a model output parameter, so as to obtain a robustness feature information determination model.
After the path feature information of the historical path is acquired in step 401 and the robustness feature information corresponding to the historical path is acquired in step 402, respectively, the path feature information of the historical path may be used as a model input parameter, the robustness feature information corresponding to the historical path may be used as a model output parameter, and the neural network model may be trained to obtain the robustness feature information determination model. The training can be performed by referring to the existing method for training the neural network model, and details are not repeated here.
Thus, the trained robustness characteristic information determination model can be obtained. Then, the path feature information of the planned path may be input into the robustness feature information determination model according to the manner shown in fig. 3, that is, the robustness feature information corresponding to the planned path may be obtained.
The following describes in detail the robustness characteristic information provided by the present disclosure, which corresponds to the planned path determined according to another non-machine learning manner. As shown in fig. 7, step 103 in fig. 1 may include steps 1033 through 1035.
In step 1033, the random distribution characteristic information is sampled according to the probability distribution of the random distribution characteristic information corresponding to the planned path. Different sampling values of the randomly distributed characteristic information are used for forming different simulation scenes of the planned path.
In step 1034, delivery time characteristic information for the planned path in each simulation scenario is determined.
Specifically, determining whether an overtime order exists in the orders related to the planned path in each simulation scene according to the estimated delivery time of each order related to the planned path in each simulation scene; aiming at each simulation scene, taking the maximum overtime time as the delivery time characteristic information of the planned path in the simulation scene under the condition that an overtime order exists in the simulation scene; and in the case that no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the planned path, and taking the average distribution margin as the delivery time characteristic information of the planned path in the simulation scene.
In step 1035, robust feature information corresponding to the planned path is determined according to the delivery time feature information of the planned path in each simulation scenario.
It should be noted that, in the present disclosure, a specific implementation manner of determining the robust feature information corresponding to the planned path is similar to the specific implementation manner of obtaining the robust feature information corresponding to the historical path, and therefore, reference may be made to the specific description of obtaining the robust feature information corresponding to the historical path in fig. 5 and fig. 6, which is not repeated herein in detail.
In addition, when the robustness feature information corresponding to the planned path is determined on line through the delivery time feature information under different simulation scenes, if the number of the simulation scenes is large, the calculation amount is too large, so that the efficiency of determining the robustness feature information corresponding to the planned path is low, and therefore, the sampling number on a control line is required in the disclosure. Specifically, the embodiment of sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path may be: and sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path and the target sampling number. The target sampling quantity is the optimal sampling quantity determined according to the probability distribution of the random distribution characteristic information corresponding to the historical path and the robustness characteristic information corresponding to the historical path under each preset sampling quantity from a plurality of preset sampling quantities.
Specifically, according to the probability distribution of the random distribution characteristic information corresponding to the historical path, sampling is sequentially carried out on the random distribution characteristic information according to different preset sampling numbers; after each sampling, determining robustness characteristic information corresponding to the historical paths under the current sampling quantity according to different simulation scenes of the historical paths formed at this time so as to obtain the corresponding relation between the preset sampling quantity and the robustness characteristic information; and taking the sampling number corresponding to the optimal robustness characteristic information as the target sampling number according to the corresponding relation.
By adopting the mode, the target sampling number is predetermined on line, and the random distribution characteristic information is sampled on line according to the probability distribution of the random distribution characteristic information corresponding to the planned path and the target sampling number. Therefore, the robustness characteristic information corresponding to the planned path can be accurately determined according to a proper number of simulated scenes, the sampling number can be reduced, the workload of on-line calculation is reduced, and the efficiency of determining the robustness characteristic information corresponding to the planned path is improved.
In addition, as described above, the robustness characteristic information corresponding to the planned path cannot be accurately determined by the machine learning method when the planned path is abnormal, and therefore, in the disclosure, the robustness characteristic information corresponding to the planned path may be determined by the machine learning method when the planned path is normal, and the robustness characteristic information corresponding to the planned path may be determined by the machine learning method when the planned path is abnormal, that is, the random distribution characteristic information corresponding to the planned path belongs to the preset abnormal information, or the environment information corresponding to the planned path belongs to the preset abnormal environment information, and the robustness characteristic information corresponding to the planned path is determined by the method shown in fig. 7. The abnormal information and the abnormal environment information are already described above, and are not described herein again.
By adopting the technical scheme, under the condition that the planned path is normal, robustness characteristic information corresponding to the planned path is determined in a machine learning mode, and under the condition that the planned path is abnormal, the robustness characteristic information corresponding to the planned path is determined through the simulation scene and according to the delivery time characteristic information under the simulation scene, so that the accuracy of determining the robustness characteristic information corresponding to the planned path is improved, and the flexibility of determining the robustness characteristic information corresponding to the planned path is also improved.
Based on the same inventive concept, the invention also provides an order distribution device. FIG. 8 is a block diagram illustrating an order distribution apparatus according to an exemplary embodiment. As shown in fig. 8, the apparatus 80 may include:
a first determination module 801 configured to determine a plurality of delivery capacities associated with an order to be allocated;
a planning module 802, configured to perform task point path planning on each of the delivery capacities respectively according to the order to be allocated and the current delivery task of each of the delivery capacities, so as to obtain a respective planned path of each of the delivery capacities;
a second determining module 803, configured to determine, for each planned path, robustness characteristic information corresponding to the planned path according to a probability distribution of random distribution characteristic information corresponding to the planned path, where the robustness characteristic information is used to characterize a degree of influence of the order to be allocated on delivery time of distribution capacity corresponding to the planned path;
a third determining module 804 configured to determine an optimal planned path according to the robustness characteristic information;
an allocating module 805 configured to allocate the order to be allocated to the delivery capacity corresponding to the optimal planned path.
Optionally, the random distribution characteristic information includes at least one of: article preparation completion time; the fetching completion time; the travel time of the delivery capacity; item delivery time.
Optionally, the second determining module includes:
a first obtaining sub-module, configured to obtain path feature information of the planned path, where the path feature information of the planned path includes a distribution parameter of a probability distribution of random distribution feature information corresponding to the planned path;
and the second obtaining submodule is configured to input the path characteristic information of the planned path to a robustness characteristic information determination model, and obtain the robustness characteristic information corresponding to the planned path.
Optionally, the path feature information further includes: the number of task points in the path and/or sequence indication information for indicating the sequence of the task points in the path.
Optionally, the apparatus further comprises:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is configured to be used for acquiring path characteristic information of a historical path, and the path characteristic information of the historical path comprises a distribution parameter of probability distribution of random distribution characteristic information corresponding to the historical path;
the second acquisition module is configured to acquire robustness characteristic information corresponding to the historical path;
and the training module is configured to train a neural network model by taking the path characteristic information of the historical path as a model input parameter and taking the robustness characteristic information corresponding to the historical path as a model output parameter so as to obtain the robustness characteristic information determination model.
Optionally, the second obtaining module includes:
a first sampling sub-module, configured to sample the random distribution feature information according to a probability distribution of the random distribution feature information included in the path feature information of the historical path, where different sampling values of the random distribution feature information are used to form different simulation scenarios of the historical path;
a first determining sub-module configured to determine delivery time characteristic information of the historical path in each simulation scenario;
and the second determining submodule is configured to determine robustness characteristic information corresponding to the historical path according to the delivery time characteristic information of the historical path in each simulated scene.
Optionally, the first determining submodule is configured to:
determining whether overtime orders exist in the orders related to the historical path under each simulation scene according to the estimated delivery time of each order related to the historical path under each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the historical path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the historical path, and taking the average distribution margin as the delivery time characteristic information of the historical path in the simulation scene.
Optionally, the apparatus further comprises:
the fourth determining module is configured to determine that the random distribution characteristic information corresponding to the planned path does not belong to preset abnormal information, and the environment information corresponding to the planned path does not belong to preset abnormal environment information.
Optionally, the second determining module includes:
the second sampling submodule is configured to sample the random distribution characteristic information according to probability distribution of the random distribution characteristic information corresponding to the planned path, wherein different sampling values of the random distribution characteristic information are used for forming different simulation scenes of the planned path;
a third determining submodule configured to determine delivery time characteristic information of the planned path in each simulation scenario;
and the fourth determining submodule is configured to determine robustness characteristic information corresponding to the planned path according to the delivery time characteristic information of the planned path in each simulated scene.
Optionally, the third determining submodule is configured to:
determining whether an overtime order exists in the orders related to the planned path in each simulation scene according to the estimated delivery time of each order related to the planned path in each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the planned path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the planned path, and taking the average distribution margin as the delivery time characteristic information of the planned path in the simulation scene.
Optionally, the second sampling sub-module is configured to:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path and a target sampling number, wherein the target sampling number is an optimal sampling number determined from a plurality of preset sampling numbers according to the probability distribution of the random distribution characteristic information corresponding to a historical path and robustness characteristic information corresponding to the historical path under each preset sampling number.
Optionally, the apparatus further comprises:
a fifth determining module, configured to determine that the random distribution feature information corresponding to the planned path belongs to preset abnormal information, or determine that the environment information corresponding to the planned path belongs to preset abnormal environment information.
Optionally, the robustness characteristic information includes at least one of: an expectation of the delivery time characteristic information for each of the simulated scenarios, a variance of the delivery time characteristic information for each of the simulated scenarios, and an expectation of the delivery time characteristic information for the simulated scenarios where there is a time-out order.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 9 is a block diagram illustrating an electronic device 900 in accordance with an example embodiment. For example, the electronic device 900 may be provided as a server. Referring to fig. 9, the electronic device 900 includes a processor 922, which may be one or more in number, and a memory 932 for storing computer programs executable by the processor 922. The computer programs stored in memory 932 may include one or more modules that each correspond to a set of instructions. Further, the processor 922 may be configured to execute the computer program to perform the order allocation method described above.
Additionally, the electronic device 900 may also include a power component 926 and a communication component 950, the power component 926 may be configured to perform power management of the electronic device 900, and the communication component 950 may be configured to enable communication, e.g., wired or wireless communication, of the electronic device 900. The electronic device 900 may also include input/output (I/O) interfaces 958. The electronic device 900 may operate based on an operating system stored in the memory 932, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, and the like.
In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the order allocation method described above is also provided. For example, the computer readable storage medium may be the memory 932 described above including program instructions that are executable by the processor 922 of the electronic device 900 to perform the order allocation method described above.
In another exemplary embodiment, a computer program product is also provided, which comprises a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-mentioned order allocation method when executed by the programmable apparatus.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, various possible combinations will not be separately described in this disclosure.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.

Claims (16)

1. An order allocation method, comprising:
determining a plurality of delivery capacities associated with the orders to be distributed;
respectively planning a task point path for each distribution capacity according to the order to be distributed and the current distribution task of each distribution capacity to obtain a respective planned path of each distribution capacity;
for each planned path, determining robustness characteristic information corresponding to the planned path according to the probability distribution of the random distribution characteristic information corresponding to the planned path, wherein the robustness characteristic information is used for representing the influence degree of the order to be distributed on the delivery time of the distribution capacity corresponding to the planned path;
determining an optimal planning path according to the robustness characteristic information;
and distributing the order to be distributed to the distribution capacity corresponding to the optimal planning path.
2. The method of claim 1, wherein the random dispatch characterization information includes at least one of: article preparation completion time; the fetching completion time; the travel time of the delivery capacity; item delivery time.
3. The method according to claim 1, wherein the determining the robustness characteristic information corresponding to the planned path according to the probability distribution of the random distribution characteristic information corresponding to the planned path comprises:
acquiring path characteristic information of the planned path, wherein the path characteristic information of the planned path comprises distribution parameters of probability distribution of random distribution characteristic information corresponding to the planned path;
inputting the path characteristic information of the planned path into a robustness characteristic information determination model to obtain the robustness characteristic information corresponding to the planned path.
4. The method of claim 3, wherein the path characteristic information further comprises: the number of task points in the path and/or sequence indication information for indicating the sequence of the task points in the path.
5. The method of claim 3, wherein the robust feature information determination model is obtained by:
acquiring path characteristic information of a historical path, wherein the path characteristic information of the historical path comprises a distribution parameter of probability distribution of random distribution characteristic information corresponding to the historical path;
acquiring robustness characteristic information corresponding to the historical path;
and training a neural network model by taking the path characteristic information of the historical path as a model input parameter and taking the robustness characteristic information corresponding to the historical path as a model output parameter so as to obtain the robustness characteristic information determination model.
6. The method according to claim 5, wherein the obtaining robustness feature information corresponding to the historical path comprises:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information included in the path characteristic information of the historical path, wherein different sampling values of the random distribution characteristic information are used for forming different simulation scenes of the historical path;
determining delivery time characteristic information of the historical path in each simulation scene;
and determining robustness characteristic information corresponding to the historical path according to the delivery time characteristic information of the historical path in each simulated scene.
7. The method of claim 6, wherein determining delivery time characterization information for the historical path in each simulation scenario comprises:
determining whether overtime orders exist in the orders related to the historical path under each simulation scene according to the estimated delivery time of each order related to the historical path under each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the historical path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the historical path, and taking the average distribution margin as the delivery time characteristic information of the historical path in the simulation scene.
8. The method of claim 3, wherein prior to the step of obtaining path characteristic information for the planned path, the method further comprises:
and determining that the random distribution characteristic information corresponding to the planned path does not belong to preset abnormal information, and determining that the environment information corresponding to the planned path does not belong to preset abnormal environment information.
9. The method according to claim 1, wherein the determining the robustness characteristic information corresponding to the planned path according to the probability distribution of the random distribution characteristic information corresponding to the planned path comprises:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path, wherein different sampling values of the random distribution characteristic information are used for forming different simulation scenes of the planned path;
determining delivery time characteristic information of the planned path in each simulation scene;
and determining robustness characteristic information corresponding to the planned path according to the delivery time characteristic information of the planned path in each simulated scene.
10. The method of claim 9, wherein determining delivery time characteristic information for the planned path in each simulation scenario comprises:
determining whether an overtime order exists in the orders related to the planned path in each simulation scene according to the estimated delivery time of each order related to the planned path in each simulation scene;
for each simulation scene, when an overtime order exists in the simulation scene, taking the maximum overtime time as the delivery time characteristic information of the planned path in the simulation scene; and when no overtime order exists in the simulation scene, determining the distribution margin of each order involved in the planned path, and taking the average distribution margin as the delivery time characteristic information of the planned path in the simulation scene.
11. The method according to claim 9 or 10, wherein the sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path includes:
sampling the random distribution characteristic information according to the probability distribution of the random distribution characteristic information corresponding to the planned path and a target sampling number, wherein the target sampling number is an optimal sampling number determined from a plurality of preset sampling numbers according to the probability distribution of the random distribution characteristic information corresponding to a historical path and robustness characteristic information corresponding to the historical path under each preset sampling number.
12. The method according to claim 9, wherein before the step of sampling the random delivery characteristic information according to the probability distribution of the random delivery characteristic information corresponding to the planned path, the method further comprises:
and determining that the random distribution characteristic information corresponding to the planned path belongs to preset abnormal information, or determining that the environment information corresponding to the planned path belongs to preset abnormal environment information.
13. The method according to claim 6 or 9, wherein the robustness feature information comprises at least one of: an expectation of the delivery time characteristic information for each of the simulated scenarios, a variance of the delivery time characteristic information for each of the simulated scenarios, and an expectation of the delivery time characteristic information for the simulated scenarios where there is a time-out order.
14. An order distribution apparatus, comprising:
a first determination module configured to determine a plurality of delivery capacities associated with an order to be allocated;
the planning module is configured to perform task point path planning on each delivery capacity respectively according to the order to be distributed and the current delivery task of each delivery capacity to obtain a respective planned path of each delivery capacity;
a second determining module, configured to determine, for each planned path, robustness characteristic information corresponding to the planned path according to a probability distribution of random delivery characteristic information corresponding to the planned path, where the robustness characteristic information is used to characterize a degree of influence of the order to be allocated on delivery time of delivery capacity corresponding to the planned path;
a third determining module configured to determine an optimal planned path according to the robustness characteristic information;
and the distribution module is configured to distribute the orders to be distributed to the distribution capacity corresponding to the optimal planning path.
15. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 13.
16. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to carry out the steps of the method of any one of claims 1 to 13.
CN202010292165.6A 2020-04-14 2020-04-14 Order distribution method, order distribution device, readable storage medium and electronic equipment Pending CN113537853A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010292165.6A CN113537853A (en) 2020-04-14 2020-04-14 Order distribution method, order distribution device, readable storage medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010292165.6A CN113537853A (en) 2020-04-14 2020-04-14 Order distribution method, order distribution device, readable storage medium and electronic equipment

Publications (1)

Publication Number Publication Date
CN113537853A true CN113537853A (en) 2021-10-22

Family

ID=78120004

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010292165.6A Pending CN113537853A (en) 2020-04-14 2020-04-14 Order distribution method, order distribution device, readable storage medium and electronic equipment

Country Status (1)

Country Link
CN (1) CN113537853A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115439071A (en) * 2022-11-09 2022-12-06 成都运荔枝科技有限公司 Cold-chain logistics transportation order processing method and system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115439071A (en) * 2022-11-09 2022-12-06 成都运荔枝科技有限公司 Cold-chain logistics transportation order processing method and system

Similar Documents

Publication Publication Date Title
CN109615201B (en) Order allocation method and device, electronic equipment and storage medium
WO2016035091A1 (en) Dynamic forecasting for forward reservation of cab
CN108806302A (en) A kind of vehicle dispatching method and device
CN110135815B (en) Travel order monitoring method and device, computer equipment and storage medium
CN112036696A (en) Task allocation method, task allocation device, storage medium, and electronic apparatus
CN110472910A (en) Determine the method, apparatus and storage medium, electronic equipment of target dispatching task node
CN111192090A (en) Seat allocation method and device for flight, storage medium and electronic equipment
CN111369137A (en) Distribution method, distribution device, server and storage medium of distribution tasks
CN109934537A (en) Order allocation method, device, server and storage medium
CN111539780B (en) Task processing method and device, storage medium and electronic equipment
CN113537853A (en) Order distribution method, order distribution device, readable storage medium and electronic equipment
CN111582407B (en) Task processing method and device, readable storage medium and electronic equipment
CN109242391A (en) A kind of cargo recognition methods and device
CN111144673A (en) Method, device and equipment for evaluating structure of organization personnel and computer readable medium
CN110689177B (en) Method and device for predicting order preparation time, electronic equipment and storage medium
CN116187675A (en) Task allocation method, device, equipment and storage medium
CN109583749A (en) A kind of software development cost intelligent control method and system based on Dynamic Programming
CN115345663A (en) Marketing strategy evaluation method and device, electronic equipment and storage medium
CN113705850A (en) Order distribution method and device, readable storage medium and electronic equipment
CN113822609A (en) Logistics line generation method and device and server
CN111105284B (en) Order processing method and device, M-layer order processing model, electronic equipment and storage medium
CN111080393A (en) Transaction matching method and device
CN112328769A (en) Automatic customer service response method, device and computer readable storage medium
CN111047354A (en) Time-sharing pricing implementation method and device, electronic equipment and storage medium
CN112819266B (en) Distribution parameter adjusting method and device, storage medium and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20211022

WD01 Invention patent application deemed withdrawn after publication