WO2021228198A1 - 一种运单分配方法、装置、存储介质和电子设备 - Google Patents

一种运单分配方法、装置、存储介质和电子设备 Download PDF

Info

Publication number
WO2021228198A1
WO2021228198A1 PCT/CN2021/093619 CN2021093619W WO2021228198A1 WO 2021228198 A1 WO2021228198 A1 WO 2021228198A1 CN 2021093619 W CN2021093619 W CN 2021093619W WO 2021228198 A1 WO2021228198 A1 WO 2021228198A1
Authority
WO
WIPO (PCT)
Prior art keywords
waybill
package
target
plan
distribution
Prior art date
Application number
PCT/CN2021/093619
Other languages
English (en)
French (fr)
Inventor
于洋
张涛
吴卓林
易琴
Original Assignee
北京三快在线科技有限公司
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 北京三快在线科技有限公司 filed Critical 北京三快在线科技有限公司
Publication of WO2021228198A1 publication Critical patent/WO2021228198A1/zh

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
    • 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
    • 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • 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
    • 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
    • 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

Definitions

  • the present disclosure relates to the field of information management, in particular, to the distribution of waybills.
  • the multi-contracting plan is an important part of the dispatch engine for cargo transportation. Specifically, the multi-contracting plan refers to a scheduling mode in which several waybills are packaged and combined before the waybill is assigned and assigned to the distribution capacity.
  • the delivery capacity can also be referred to as delivery resources, including but not limited to delivery personnel (takeaway riders, couriers, etc.) and delivery vehicles. In the following, “resources” and “capacity” are used interchangeably.
  • the waybills to be distributed can be packaged and combined in advance to form a waybill package, and then these waybill packages are dispatched to the distribution capacity through the distribution algorithm.
  • the basic principle of the waybill package distribution is to improve the delivery efficiency of the delivery staff while reducing the delivery cost of the takeaway or express platform. Specifically, it is to increase the degree of delivery of each waybill in the waybill package, so as to increase the number of waybills delivered per unit time, and based on the improvement of the degree of delivery, a certain degree of deduction will be made on the delivery fee paid to the delivery capacity. , Reduce distribution costs.
  • the main purpose of the present disclosure is to provide a waybill distribution method, device, storage medium, and electronic equipment, so as to solve the technical problem that the accuracy of the waybill package is low in the related art, which affects the efficiency of the waybill distribution.
  • the first aspect of the present disclosure provides a waybill distribution method, the method includes:
  • At least one combined plan is formulated for multiple target waybills, wherein the acceptance is the acceptance determined based on the acceptance behavior of the historical waybill package by the distribution resource, and the combined plan Used to merge the multiple target waybills into one or more target waybill packages;
  • each target waybill package determine the target combined package plan from the at least one combined package plan
  • the formulating at least one combined package plan for multiple target waybills based on the acceptance of the distribution resources to the historical waybill package includes:
  • each waybill package group includes one or more waybill packages obtained by merging the multiple target waybills;
  • one or more target waybill package groups are determined, and one or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • the acceptance mark includes a first mark and a second mark, the first mark is used to indicate that the waybill package is accepted by the distribution resource, and the second mark is used to indicate that the waybill package is not accepted by the distribution resource.
  • the acceptance mark corresponding to each waybill package one or more target waybill package groups are determined, so that one or more target waybill packages included in the target waybill package group are used as the combined package plan, including :
  • One or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • the acceptance prediction model is obtained in the following manner:
  • the preset prediction model is trained through the training data to generate the acceptance prediction model.
  • the feature data of the waybill package includes one or more of the following features: waybill feature, waybill package feature, and waybill area feature, where the waybill feature includes the distribution data of the delivery process of each waybill in the waybill package
  • the feature of the waybill package is used to characterize the relationship between multiple waybills in the waybill package
  • the feature of the waybill area is used to characterize the distribution data of the distribution area corresponding to the waybill package.
  • the determining the target combined package plan from the at least one combined package plan according to the delivery time length of each target waybill package includes:
  • For each combined plan determine the delivery time of each target waybill package included in the combined plan, where the delivery time is the time required to complete the delivery of all target waybills in the target waybill package;
  • the combined package plan with the shortest total delivery time is taken as the target combined package plan.
  • the allocating the target waybill package corresponding to the target combined plan to the distribution resource includes:
  • the status information includes: time information and location information of the distribution task currently assigned to each distribution resource;
  • All target waybills contained in the target waybill package corresponding to the target distribution resource are allocated to the target distribution resource.
  • a second aspect of the present disclosure provides a waybill distribution device, which includes:
  • the plan formulation module is configured to formulate at least one combined plan for multiple target waybills according to the acceptance of the distribution resources to the historical waybill package, wherein the acceptance is determined based on the acceptance behavior of the historical waybill package by the distribution resource
  • the acceptance degree of, the combined package scheme is used to merge the multiple target waybills into one or more target waybill package resources;
  • the plan determination module is configured to determine a target combined package plan from the at least one combined package according to the delivery time of each target waybill package;
  • the waybill distribution module is configured to allocate one or more target waybill packages corresponding to the target combined plan to distribution resources according to the target combined plan.
  • the solution formulation module is configured to:
  • each waybill package group includes one or more waybill packages obtained by merging the multiple target waybills;
  • one or more target waybill package groups are determined, and one or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • the acceptance mark includes a first mark and a second mark
  • the first mark is used to indicate that the waybill package is accepted by the distribution resource
  • the second mark is used to indicate that the waybill package is not accepted by the distribution resource.
  • One or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • the acceptance prediction model is obtained in the following manner:
  • the preset prediction model is trained through the training data to generate an acceptance prediction model.
  • the feature data of the waybill package includes one or more of the following features: waybill feature, waybill package feature, and waybill area feature.
  • the waybill feature includes the distribution data of the delivery process of each waybill in the waybill package, so
  • the waybill package feature is used to characterize the relationship between multiple waybills in the waybill package, and the waybill area feature is used to characterize the distribution data of the distribution area corresponding to the waybill package.
  • the solution determining module is configured to:
  • For each combined plan determine the delivery time of each target waybill package included in the combined plan, where the delivery time is the length of time required to complete the delivery of all target waybills in the target waybill package;
  • the combined package plan with the shortest total delivery time is taken as the target combined package plan.
  • the waybill distribution module is configured to:
  • the status information includes: time information and location information of the distribution task currently assigned to each distribution resource;
  • All target waybills contained in the target waybill package corresponding to the target distribution resource are allocated to the target distribution resource.
  • a third aspect of the present disclosure provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the waybill distribution method described in the first aspect are implemented.
  • a fourth aspect of the present disclosure provides an electronic device, including:
  • the processor is configured to execute the computer program in the memory to implement the waybill distribution method described in the first aspect.
  • At least one combined package plan is developed for multiple target waybills, where the combined package plan is used to merge the multiple target waybills into one or more target waybill packages, and each target All target waybills in the waybill package are assigned to the same delivery capacity.
  • the acceptance is based on the acceptance of the historical waybill package by the distribution capacity; according to the delivery time of each target waybill package, from the above at least one package plan Determine the target package plan; assign the target waybill package corresponding to the target package plan to the distribution resources.
  • the combined package plan can be determined, so as to improve the fit between the waybill package and the willingness of the distribution resource, thereby improving the efficiency of the waybill distribution.
  • Fig. 1 is a flow chart showing a method for allocating waybills according to an exemplary embodiment
  • FIG. 2 is a flowchart of another way of allocating a waybill shown in FIG. 1;
  • Fig. 3 is a flowchart of a method for formulating a contract plan according to Fig. 2;
  • Fig. 4 is a flowchart of a method for determining a package plan according to Fig. 2;
  • Fig. 5 is a flowchart of a method for allocating waybill packages shown in Fig. 2;
  • Fig. 6 is a block diagram showing a waybill distribution device according to an exemplary embodiment
  • FIG. 7 is a block diagram of another waybill distribution device shown in FIG. 6;
  • Fig. 8 is a schematic structural diagram showing an electronic device according to an exemplary embodiment.
  • the above-mentioned co-packing algorithm does not consider the acceptance of the distribution capacity to the assigned waybill package, but relies on manual means outside the algorithm to analyze the distribution capacity's acceptance behavior data on the assigned waybill package, and obtain Acceptance data, and then use acceptance data as reference data to maintain and iterate the merge conditions in the package algorithm through manual means, failing to realize the automatic perception of the dispatching system’s delivery capacity behavior, thereby reducing the cost of package and waybill allocation efficient.
  • the present disclosure proposes a waybill distribution method, which is specifically as follows.
  • Fig. 1 is a flow chart showing a method for allocating waybills according to an exemplary embodiment. As shown in Fig. 1, the method includes step 101 to step 103.
  • Step 101 According to the acceptance of the historical waybill package by the distribution capacity, at least one combined package plan is formulated for multiple target waybills.
  • the waybills contained in each waybill package are assigned to the same distribution capacity.
  • the combined package scheme is used to merge the above multiple target waybills into one or more target waybill packages.
  • the acceptance is based on the historical waybill package The acceptance behavior determines the degree of acceptance.
  • the delivery capacity can be delivery personnel for takeout or express delivery, or delivery equipment such as drones and unmanned vehicles for the delivery of goods.
  • the target waybill may be a takeaway waybill that needs to be delivered or an express delivery waybill. Take the takeaway waybill as an example.
  • the takeaway waybill is not directly allocated to the delivery capacity at the first time, but after a certain amount is accumulated, the package will be processed, and then the waybill will be packaged. Make an assignment.
  • one of the multi-packing plan can be: Combine the waybills numbered 1-3 into one The waybill package, the waybills numbered 4-9 are combined into one waybill package, and the waybill number 10 is a single waybill package. It should be noted that in this step 101, after receiving the above multiple target waybills, at least one combined package plan can be determined through different combined package algorithms, and each combined package plan contains all target waybills, but the target waybill The way of combining into a waybill package is not the same.
  • the acceptance behavior of the historical waybill package by the distribution capacity includes: the distribution capacity has accepted the distribution of the historical waybill package, and the distribution capacity has refused the distribution of the historical waybill package.
  • Step 102 according to the delivery time of each target waybill package, determine the target combined package plan from the above at least one combined package plan.
  • the delivery duration is the total duration of delivery of the target waybill package through the optimal route. It is understandable that after the package is combined, the distribution capacity will be distributed on the waybills in units of each waybill package. In this case, the shorter the distribution time, the higher the efficiency of the distribution capacity.
  • the basic principle of distribution of takeaway or express waybills is to improve the delivery efficiency of delivery personnel and reduce the delivery cost of takeaway or express delivery platforms.
  • improving the delivery efficiency lies in reducing the delivery time and reducing the final rate of the waybills assigned by the delivery personnel, and reducing the cost lies in minimizing the postage and other expenses spent by the delivery personnel when the delivery personnel complete the delivery tasks.
  • the functional positioning of the combined package in the entire scheduling system in the embodiments of the present disclosure is also to reduce the distribution cost on the basis of taking into account the distribution efficiency.
  • the principle of consolidating the waybill to reduce the cost is as follows: improving the delivery efficiency of a single delivery journey by the delivery personnel, and at the same time discounting the postage according to the order of the waybill package. Improving the delivery efficiency of a single delivery itinerary of delivery personnel means that the income of delivery personnel per unit time of this delivery journey will increase, and the deduction of postage according to the degree of order will reduce the cost of the platform to the delivery staff during this delivery itinerary.
  • the specific pricing rules can be as shown in the following formula (1):
  • T1 is the total time for the combined delivery of the waybill package using the optimal route
  • T2 is the time for each waybill to be delivered individually
  • n is the number of waybills included in the waybill package.
  • formula (1) It is a mathematical description of the relationship between orderliness and postage discounts. It should be noted that the greater the value of the discount, the smaller the strength of the discount, and the smaller the value of the discount, the greater the strength of the discount.
  • discount_0 is the pre-set minimum postage discount, which is used to limit the discount value to not too small to protect the interests of the delivery personnel.
  • the discount_0 can be set to 0.5 (that is, 50% of the default postage is paid to the delivery staff), if The calculation result of is 0.1 (that is, 10% of the preset postage is paid to the delivery personnel), then the larger value, namely 0.5, is used for postage deduction.
  • the delivery time of each waybill package corresponding to each package plan needs to be calculated, and the delivery time of the package plan should be calculated based on the delivery time of the package plan.
  • These package plans are further screened to determine the target package plans.
  • Step 103 Allocate one or more target waybill packages corresponding to the target combined package to the distribution capacity.
  • all target waybill packages in the target combined package plan can be allocated to the corresponding distribution based on the actual addresses and tasks of all delivery capacity in a region and/or a period of time Capacity.
  • each target waybill package received by the delivery capacity may not only contain the target waybill to be distributed, but also a distribution plan for all the target waybills in each target waybill package.
  • the delivery plan can include the pickup and delivery order of all target waybills.
  • the technical solutions provided by the embodiments of the present disclosure can formulate at least one package plan for multiple target waybills according to the acceptance of the distribution capacity to the historical waybill package.
  • Multiple target waybills are merged into one or more target waybill packages, and all target waybills in each target waybill package are assigned to the same delivery capacity.
  • the acceptance is the acceptance based on the acceptance behavior of the historical waybill package based on the delivery capacity;
  • According to the delivery time of each target waybill package determine the target consignment plan from the above at least one consignment plan; assign the target consignment package corresponding to the target consignment plan to the distribution capacity.
  • Fig. 2 is a flowchart of another waybill distribution method shown in Fig. 1.
  • the acceptance of the distribution capacity to the historical waybill package can be determined through a preset acceptance prediction model Therefore, before step 101, the method further includes a training process of the acceptance prediction model, and the training process may include the following steps.
  • Step 104 Extract feature data of multiple historical waybill packages and the acceptance mark corresponding to each historical waybill package as training data.
  • the acceptance mark is used to characterize the acceptance of the distribution capacity to the historical waybill package.
  • the feature data includes one or more of the following features: waybill feature, waybill package feature, and waybill area feature.
  • the waybill feature includes the distribution data of the distribution process of each waybill in the waybill package.
  • the waybill package feature is used to characterize the waybill.
  • the relationship between the multiple waybills in the package, the waybill area feature is used to characterize the overall distribution data of the distribution area corresponding to the waybill package, and the acceptance mark is used to characterize whether the waybill package has been accepted by the distribution capacity.
  • the acceptance is divided into duality, that is, the acceptance and rejection behavior of the distribution capacity for the allocated historical waybill package is represented by the acceptance mark (or label).
  • the acceptance mark corresponding to the waybill package includes a first mark and a second mark. The first mark is used to indicate that the waybill package is accepted by the distribution capacity, and the second mark is used to indicate that the waybill package is not accepted by the distribution capacity. Based on this, the prediction of the acceptance behavior of the delivery capacity to the waybill package can be regarded as a classification problem.
  • the process can be: train the preset classification model (or prediction model) through the historical waybill package’s waybill characteristics, waybill package characteristics, and waybill area characteristics, as well as the acceptance mark of each historical waybill package, and then pass the training.
  • the classification model predicts the current waybill package.
  • the first mark may be expressed as 1
  • the second mark may be expressed as 0.
  • the aforementioned characteristics of the waybill may include: the meal time of each waybill in the waybill package, the agreed delivery time, and so on.
  • the features of the above-mentioned waybill package may include: the distance between the pick-up points of each waybill in the waybill package and the co-ordination coefficient after the package.
  • the aforementioned regional characteristics of the waybill may include: current environmental characteristics and historical statistical characteristics of the area where the waybill is located.
  • the current environmental characteristics may include: the load status of the regional air waybill, the regional weather level, and the current unaccepted volume of the region, etc.
  • the historical statistical characteristics may include the waiting time for meals in the area within a preset time period before the current time point, the probability of rejection of the waybill, and the average area covered by the delivery route of the delivery capacity in the area.
  • the training data can also be other data that can reflect the characteristics of the waybill, waybill package, and delivery area as complete as possible.
  • the feature type of the training data ie, the historical waybill package
  • the feature type of the waybill package is the same as the feature type of the waybill package to be predicted.
  • Step 105 Train a preset prediction model through the training data to generate an acceptance prediction model.
  • step 104 and step 105 a large amount of feature data can be collected for one-time training of the prediction model, or model training tasks can be customized in the system to periodically update the acceptance prediction model .
  • the model training task may be, for each preset duration, for example, one day or one week, collecting feature data and acceptance marks of the waybill packets processed within the preset duration, and retraining and retraining the acceptance prediction model. Update to ensure that the acceptance prediction model is sensitive to environmental changes in the time and area.
  • FIG. 3 is a flowchart of a method for formulating a package plan according to FIG. 2. As shown in FIG. 3, this step 101 includes step 1011 to step 1014.
  • Step 1011 Determine one or more waybill package groups corresponding to the multiple target waybills.
  • each waybill package group includes one or more waybill packages obtained by merging the multiple target waybills.
  • the above multiple target waybills can be grouped by a variety of grouping rules
  • the grouping rules include but are not limited to: grouping rules based on delivery location, grouping rules based on delivery time, and determination based on a preset clustering model Grouping rules, etc.
  • Step 1012 For each waybill package group, obtain characteristic data of each waybill package in the waybill package group.
  • Step 1013 Obtain the acceptance mark corresponding to each of the above-mentioned waybill packages according to the characteristic data of each of the above-mentioned waybill packages and the preset acceptance degree prediction model.
  • the step 1013 may include: using the characteristic data of each of the above-mentioned waybill packages as the input of the acceptance prediction model to obtain the acceptance mark corresponding to each of the above-mentioned waybill packages output by the acceptance prediction model.
  • the acceptance prediction model may be an xgboost model (extreme Gradient Boosting Tree), an SVM model (Support Vector Machine) or a neural network model that has been pre-trained in the above steps 104 and 105.
  • the characteristic data of each waybill package in each waybill package group is used as the input of the acceptance prediction model, that is, the acceptance mark corresponding to each of the above-mentioned waybill packages output by the xgboost model can be obtained.
  • Step 1014 Determine one or more target waybill package groups according to the acceptance mark corresponding to each of the above-mentioned waybill packages, and use one or more target waybill packages included in the target waybill package group as the combined package plan.
  • the elements corresponding to each waybill package group are consistent with the elements corresponding to each combined package plan, and they are all the above multiple goals.
  • Waybill In the case of uncertain whether the combination of the waybill meets the conditions of the acceptance behavior, these packaged target waybills are called a waybill package group. When it is determined that the combination of the waybill meets the conditions of the acceptance behavior, these packaged target waybills This is called a combined package.
  • the acceptance behavior condition can be that all the waybill packages in a waybill package group (or the waybill packages exceeding a certain proportion) are attached with the above-mentioned first mark, that is, all are accepted by the delivery capacity, then the waybill package group can be regarded as It is a combined package.
  • the step 1014 includes: obtaining the target proportion of the first waybill package in each waybill package group, where the first waybill package is the waybill package with the first mark; and the target proportion is greater than the preset ratio threshold.
  • the waybill package group is used as the target waybill package group; one or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • the actual output of the xgboost model is the acceptance prediction value, which is positively correlated with the probability that the delivery capacity accepts the waybill package.
  • the acceptance prediction value output by the model after the threshold of the xgboost model is determined by the preset recall rate, the acceptance mark can be determined according to the threshold and the acceptance prediction value. For example, if the predicted acceptance value of the waybill package is greater than the threshold, the acceptance mark of the waybill package is considered as the first mark; if the predicted acceptance value of the waybill package is less than or equal to the threshold, the acceptance mark of the waybill package is considered The second mark.
  • the xgboost model can be backtracked based on the predicted acceptance value.
  • the probability of the delivery capacity accepting the waybill package can be determined based on the predicted acceptance value, and then the objective function of the xgboost model can be determined based on the probability Make corrections to improve the prediction accuracy of the xgboost model while the model is running.
  • Fig. 4 is a flow chart of a method for determining a package plan according to Fig. 2. As shown in Fig. 4, this step 102 includes the following steps.
  • Step 1021 For each combined plan, determine the delivery time of each target waybill package included in the combined plan.
  • the delivery time is the time required to complete the delivery of all the target waybills in the target waybill package.
  • Step 1022 Obtain the sum of the delivery time of each target waybill package included in the combined plan as the total delivery time of the combined plan.
  • Step 1023 Use the combined package plan with the shortest total delivery time as the target combined package plan.
  • step 1021 may include: after analyzing and planning the distribution path corresponding to each target waybill package through a preset path analysis algorithm, obtaining the optimal distribution path of each target waybill package, the optimal distribution path may be The path with the shortest time or the path with the shortest distance. After that, the delivery time required to deliver all the target waybills in the target waybill package through the optimal delivery route is estimated as the delivery time of the target waybill package. Based on the basic principle of improving the efficiency of distribution, it is necessary to determine the shortest average delivery time of the package from all the package plans.
  • the average delivery time of the order can be understood as the average value of the delivery time of the multiple to-be-delivered waybills, which is determined after the multiple to-be-delivered waybills are consolidated through a certain combination scheme. It is understandable that, in this embodiment, all the multi-contracting plans are formulated for the multiple target air waybills mentioned above, and no matter which multi-contracting scheme is adopted, the number of the target air waybills is certain. The total delivery time divided by the number of target waybills is the average delivery time of a multi-package plan. Since the total number of target air waybills is fixed, the package plan with the shortest total delivery time can be directly identified as the target package plan.
  • Fig. 5 is a flowchart of a method for allocating waybill packages shown in Fig. 2. As shown in Fig. 5, this step 103 includes the following steps.
  • Step 1031 Determine the target distribution capacity corresponding to each target waybill package in the target combined package from the multiple distribution capacity according to the respective status information of the multiple distribution capacity in the preset area corresponding to the multiple target waybills.
  • the status information includes: time information and location information of the distribution task to which each distribution capacity is currently allocated.
  • Step 1032 Assign all target waybills included in the target waybill package corresponding to the target distribution capacity to the target distribution capacity.
  • the waybills in the same area are grouped together for the formulation of the package plan and the subsequent distribution of the waybill.
  • the target distribution capacity that can achieve the target waybill package distribution can be determined according to the status information of each distribution capacity in the area. For example, if there is no conflict between the assigned delivery task of delivery capacity A and the time and location of the delivery task in the target waybill package B, the delivery capacity A can be used as the target delivery capacity corresponding to the target waybill package B .
  • the target delivery capacity is repeatedly determined for each target waybill package in the target combined plan until each of the above-mentioned waybill packages corresponds to the target delivery capacity.
  • the waybill distribution can be performed according to the corresponding relationship between the target waybill package and the target delivery capacity. It is understandable that each target waybill package is allocated as a whole, that is, all target waybills contained in the target waybill package are allocated to their corresponding target distribution capacity.
  • the embodiments of the present disclosure it is possible to formulate at least one combined package plan for multiple target air waybills according to the acceptance of the distribution capacity to the historical waybill package, wherein the combined package solution is used to combine the multiple targets mentioned above.
  • the waybills are merged into one or more target waybill packages, and all target waybills in each target waybill package are assigned to the same delivery capacity.
  • the acceptance is the acceptance based on the acceptance behavior of the historical waybill package based on the delivery capacity; For the delivery time of the target air waybill package, determine the target multi-packing plan from the above at least one multi-packing plan; allocate the target air waybill package corresponding to the target multi-packing plan to the distribution capacity.
  • Fig. 6 is a block diagram showing a waybill distribution device according to an exemplary embodiment. As shown in Fig. 6, the device 600 includes:
  • the plan formulation module 610 is configured to formulate at least one combined plan for multiple target waybills according to the acceptance of the historical waybill package by the distribution capacity. Degree, the combined package plan is used to merge the above multiple target waybills into one or more target waybill packages;
  • the plan determination module 620 is configured to determine a target combined package plan from the above at least one combined package according to the delivery time of each target waybill package;
  • the waybill allocation module 630 is configured to allocate one or more target waybill packages corresponding to the target combined package to the delivery capacity.
  • the solution formulation module 610 is configured to:
  • each waybill package group includes one or more waybill packages obtained by merging the multiple target waybills;
  • one or more target waybill package groups are determined, and one or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • the acceptance mark includes a first mark and a second mark, the first mark is used to indicate that the waybill package is accepted by the distribution capacity, and the second mark is used to indicate that the waybill package is not accepted by the distribution capacity.
  • the plan formulation module 610 is configured to:
  • the waybill package group whose target proportion is greater than the preset ratio threshold is taken as the target waybill package group;
  • One or more target waybill packages included in the target waybill package group are used as the combined package plan.
  • FIG. 7 is a block diagram of another waybill distribution device shown in FIG. 6, which can determine the acceptance of delivery capacity for waybill packages through a preset acceptance prediction model.
  • the data extraction module 640 and the model training module 650 implement the training process of the acceptance prediction model.
  • the device 600 further includes:
  • the data extraction module 640 is configured to extract feature data of multiple historical waybill packages and the acceptance mark corresponding to each historical waybill package as training data;
  • the model training module 650 is configured to train a preset prediction model through the training data to generate an acceptance prediction model.
  • the feature data of the waybill package includes one or more of the following features: waybill feature, waybill package feature, and waybill area feature, where the waybill feature includes the distribution data of the delivery process of each waybill in the waybill package,
  • the waybill package feature is used to characterize the relationship between multiple waybills in the waybill package, and the waybill area feature is used to characterize the distribution data of the distribution area corresponding to the waybill package.
  • the solution determining module 620 is configured to:
  • For each combined plan determine the delivery time of each target waybill package included in the combined plan, which is the length of time required to complete the delivery of all target waybills in the target waybill package;
  • the combined package plan with the shortest total delivery time is taken as the target combined package plan.
  • the waybill distribution module 630 is configured to:
  • the target delivery capacity corresponding to each target waybill package in the target combined package is determined from the multiple delivery capacities, the The status information includes: time information and location information of the distribution tasks currently assigned for each distribution capacity;
  • All target waybills contained in the target waybill package corresponding to the target distribution capacity are allocated to the target distribution capacity.
  • the embodiments of the present disclosure it is possible to formulate at least one combined package plan for multiple target waybills according to the acceptance of the distribution capacity for historical waybills, where the combined package plan is used to combine the multiple target waybills. Merge into one or more target waybill packages. All target waybills in each target waybill package are assigned to the same distribution capacity. The acceptance is based on the acceptance of the historical waybill package by the distribution capacity; according to each target For the delivery time of the waybill package, determine the target package plan from the above at least one package plan; allocate the target waybill package corresponding to the target package plan to the distribution capacity.
  • Fig. 8 is a block diagram showing an electronic device 800 according to an exemplary embodiment.
  • the electronic device 800 may be provided as a server.
  • the electronic device 800 includes a processor 801, the number of which may be one or more, and a memory 802 configured to store a computer program executable by the processor 801.
  • the computer program stored in the memory 802 may include one or more modules each corresponding to a set of instructions.
  • the processor 801 may be configured to execute the computer program to execute the waybill distribution method shown in FIGS. 1 to 4 above.
  • the electronic device 800 may further include a power supply component 803 and a communication component 804.
  • the power supply component 803 may be configured to perform power management of the electronic device 800
  • the communication component 804 may be configured to implement the communication of the electronic device 800, for example, wired Or wireless communication.
  • the electronic device 800 may further include an input/output (I/O) interface 805.
  • the electronic device 800 can operate based on an operating system stored in the memory 802, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM and so on.
  • a computer-readable storage medium including program instructions that, when executed by a processor, implement the steps of the waybill distribution method shown in FIGS. 1 to 4.
  • the computer-readable storage medium may be the above-mentioned memory 802 including program instructions, which may be executed by the processor 801 of the electronic device 800 to complete the above-mentioned waybill distribution method shown in FIGS. 1 to 4.

Landscapes

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

Abstract

一种运单分配方法、装置、存储介质和电子设备，运单分配方法包括：根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，接受度为基于配送资源对历史运单包的接受行为确定的接受度，合包方案用于将多个目标运单合并为一个或多个目标运单包；根据每个目标运单包的配送时长，从至少一个合包方案中确定目标合包方案；将目标合包方案对应的目标运单包分配至配送资源。

Description

[根据细则37.2由ISA制定的发明名称]　一种运单分配方法、装置、存储介质和电子设备 技术领域
本公开涉及信息管理领域，具体地，涉及运单分配。
背景技术
随着移动互联网的普及，越来越多的人选择通过网络购买生活用品和外卖食品等实体物品。在通过网络购买实体物品的交易过程中，会涉及通过邮寄和人工运送的方式进行货物运送的过程。合包方案是货物运送的调度引擎的重要组成部分，具体地，合包方案是指在运单指派前将若干运单打包组合后分配给配送运力的一种调度模式。该配送运力也可称为配送资源，包括但不限于配送人员(外卖骑手、快递员等)以及配送车辆。在下文中，“资源”与“运力”可互换使用。在货物运送的调度场景中，在将所有待分配的运单分发给配送运力前，可以预先对待分配的运单进行打包组合以形成运单包，再将这些运单包通过配送算法调度给配送运力。运单打包分配的基本原则是提高配送人员的配送效率同时降低外卖或快递平台花费的配送成本。具体来说，就是提高对运单包中每个运单进行配送的顺路程度，以提高单位时间内配送的运单数量，而基于顺路程度的提高，对向配送运力支付的配送费用进行一定程度的扣减，降低配送成本。
发明内容
本公开的主要目的是提供一种运单分配方法、装置、存储介质和电子设备，以解决相关技术中运单合包的准确率较低进而影响运单分配的效率的技术问题。
为了实现上述目的，本公开第一方面提供一种运单分配方法，所述方法包括：
根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，所述接受度为基于配送资源对历史运单包的接受行为确定的接受度，所述合包方案用于将所述多个目标运单合并为一个或多个目标运单包；
根据每个目标运单包的配送时长，从所述至少一个合包方案中确定目标合包方案；
将所述目标合包方案对应的目标运单包分配至配送资源。
可选的，所述根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，包括：
确定所述多个目标运单对应的一个或多个运单包组，每个运单包组包括对所述多个目标运单进行合并所获得的一个或多个运单包；
针对于每个运单包组，获取该运单包组中的每个运单包的特征数据；
根据所述每个运单包的特征数据以及预设的接受度预测模型，获取所述每个运单包对应的接受度标记；
根据所述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案。
可选的，所述接受度标记包括第一标记和第二标记，所述第一标记用于表征运单包被配送资源接受，所述第二标记用于表征运单包不被配送资源接受，所述根据所述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将所述目标运单包组 中包含的一个或多个目标运单包作为所述合包方案，包括：
获取每个运单包组中第一运单包的目标占比，所述第一运单包为具备所述第一标记的运单包；
将所述目标占比大于预设比率阈值的运单包组，作为所述目标运单包组；
将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案。
可选的，所述接受度预测模型采用以下方式得到：
提取多个历史运单包的特征数据和每个历史运单包对应的接受度标记，作为训练数据；
通过所述训练数据对预设的预测模型进行训练，以生成所述接受度预测模型。
可选的，运单包的特征数据包括以下特征中的一个或多个：运单特征、运单包特征和运单区域特征，其中，所述运单特征包括运单包中的每个运单的配送过程的配送数据，所述运单包特征用于表征运单包中的多个运单之间的关系，所述运单区域特征用于表征运单包对应的配送区域的配送数据。
可选的，所述根据每个目标运单包的配送时长，从所述至少一个合包方案中确定目标合包方案，包括：
针对于每个合包方案，确定该合包方案包含的每个目标运单包的配送时长，所述配送时长为将目标运单包中的所有目标运单配送完毕所需的时长；
获取该合包方案包含的各个目标运单包的配送时长的总和，作为该合包方案的总配送耗时；
将具备最短的总配送耗时的合包方案作为所述目标合包方案。
可选的，所述将所述目标合包方案对应的目标运单包分配至配送资源，包括：
根据所述多个目标运单对应的预设区域内多个配送资源各自的状态信息，从所述多个配送资源中确定所述目标合包方案中的每个目标运单包对应的目标配送资源，所述状态信息包括：每个配送资源当前已被分配的配送任务的时间信息和位置信息；
将所述目标配送资源对应的目标运单包中包含的所有目标运单分配至所述目标配送资源。
本公开第二方面提供一种运单分配装置，所述装置包括：
方案制定模块，被配置成用于根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，所述接受度为基于配送资源对历史运单包的接受行为确定的接受度，所述合包方案用于将所述多个目标运单合并为一个或多个目标运单包资源；
方案确定模块，被配置成用于根据每个目标运单包的配送时长，从所述至少一个合包方案中确定目标合包方案；
运单分配模块，被配置成用于根据所述目标合包方案，将所述目标合包方案对应的一个或多个目标运单包分配至配送资源。
可选的，所述方案制定模块，被配置成用于：
确定所述多个目标运单对应的一个或多个运单包组，每个运单包组包括对所述多个目标运单进行合并所获得的一个或多个运单包；
针对于每个运单包组，获取该运单包组中的每个运单包的特征数据；
根据所述每个运单包的特征数据以及预设的接受度预测模型，获取所述每个运单包对应的接受度标记；
根据所述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案。
可选的，所述接受度标记包括第一标记和第二标记，所述第一标记用于表征运单包被配送资源接受，所述第二标记用于表征运单包不被配送资源接受，所述方案制定模块，被配置成用于：
获取每个运单包组中第一运单包的目标占比，所述第一运单包为具备所述第一标记的运单包；
将所述目标占比大于预设比率阈值的运单包组，作为所述目标运单包组；
将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案。
可选的，所述接受度预测模型采用以下方式得到：
提取多个历史运单包的特征数据和每个历史运单包对应的接受度标记，作为训练数据；
通过所述训练数据对预设的预测模型进行训练，以生成接受度预测模型。
可选的，运单包的特征数据以下特征中的一个或多个包括：运单特征、运单包特征和运单区域特征，所述运单特征包括运单包中的每个运单的配送过程的配送数据，所述运单包特征用于表征运单包中的多个运单之间的关系，所述运单区域特征用于表征运单包对应的配送区域的配送数据。
可选的，所述方案确定模块，被配置成用于：
针对于每个合包方案，确定该合包方案包含的每个目标运单包的配送时长，所述配送时长为将该目标运单包中的所有目标运单配送完毕所需的时长；
获取该合包方案包含的各个目标运单包的配送时长的总和，作为该合包方案的总配送耗时；
将具备最短的总配送耗时的合包方案作为所述目标合包方案。
可选的，所述运单分配模块，被配置成用于：
根据所述多个目标运单对应的预设区域内多个配送资源各自的状态信息，从所述多个配送资源中确定所述目标合包方案中的每个目标运单包对应的目标配送资源，所述状态信息包括：每个配送资源当前已被分配的配送任务的时间信息和位置信息；
将所述目标配送资源对应的目标运单包中包含的所有目标运单分配至所述目标配送资源。
本公开第三方面提供一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现第一方面所述的运单分配方法的步骤。
本公开第四方面提供一种电子设备，包括：
存储器，其上存储有计算机程序；
处理器，用于执行所述存储器中的所述计算机程序，以实现第一方面所述的运单分配方法。
采用本公开提供的技术方案，至少可以达到如下技术效果：
根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，该合包方案用于将上述多个目标运单合并为一个或多个目标运单包，每个目标运单包中的所有目标运单被分配至同一配送运力，该接受度为基于配送运力对历史运单包的接受行为确定的接受度；根据每个目标运单包的配送时长，从上述至少一个合包方案中确定目标合包方案；将该目标合包方案对应的目标运单包分配至配送资源。能够根据配送资源对于已分配的运单包的接受行为和运单包的时间成本确定合包方案，提高运单包与配送资源意愿的契合度，进而提高运单分配效率。
本公开的其他特征和优点将在随后的具体实施方式部分予以详细说明。
附图说明
附图是用来提供对本公开的进一步理解，并且构成说明书的一部分，与下面的具体实施方式一起用于解释本公开，但并不构成对本公开的限制。在附图中：
图1是根据一示例性实施例示出的一种运单分配方法的流程图；
图2是根据图1示出的另一种运单分配方法的流程图；
图3是根据图2示出的一种制定合包方案的方法的流程图；
图4是根据图2示出的一种确定合包方案的方法的流程图；
图5是根据图2示出的一种分配运单包的方法的流程图；
图6是根据一示例性实施例示出的一种运单分配装置的框图；
图7是根据图6示出的另一种运单分配装置的框图；
图8是根据一示例性实施例示出的一种电子设备的结构示意图。
具体实施方式
以下结合附图对本公开的具体实施方式进行详细说明。应当理解的是，此处所描述的具体实施方式仅用于说明和解释本公开，并不用于限制本公开。
在对运单进行合包的相关技术中，通常只对符合预设条件的运单进行组合。即，将所有运单中符合某一合并条件的运单合并为一个运单包。该合包算法中的合并条件的提出与论证受人工和环境的影响较大，难以覆盖合理合包场景，进而使得合并出的运单包不能很好地契合配送运力(例如，配送人员或无人配送设备的管理方)的意愿。因此，在原有合包算法下合包率较低，在整个调度系统中降低成本能力有限。并且，上述的合包算法并未考虑配送运力对于已分配的运单包的接受度，而是在算法之外依赖人工手段，来对配送运力关于已分配的运单包的接受行为数据进行分析，得到接受度数据，进而将接受度数据作为参考数据通过人工手段对合包算法中的合并条件进行维护和迭代，未能实现调度系统对配送运力行为的自动化感知，进而降低了合包和运单分配的效率。
对此，本公开提出了一种运单分配方法，具体如下。
图1是根据一示例性实施例示出的一种运单分配方法的流程图，如图1所示，该方法包括步骤101至步骤103。
步骤101，根据配送运力对于历史运单包的接受度，为多个目标运单制定至少一个合包方案。
其中，每个运单包中包含的运单被分配至同一配送运力，该合包方案用于将上述多个目标运单合并为一个或多个目标运单包，该接受度为根据配送运力对历史运单包的接受行为确定的接受度。该配送运力可以为外卖或快递的配送人员，或者，用于货物配送的无人机、无人驾驶车辆等配送设备。
示例地，目标运单可以为需要配送的外卖运单或者快递配送运单。以外卖运单为例，在本实施例中，每个外卖运单被提交后并不是第一时间就直接分配给配送运力，而是在积攒到一定数量后进行合包处理，再以运单包的形式进行分配。在合包的过程中，首先要确定合包方案，例如，在一段时间内收集到编号1-10的十个运单，其中一种合包方案可以为：将编号1-3的运单合并成一个运单包，编号4-9的运单合并成一个运单包，编号10的运单单独为一个运单包。需要说明的是，在该步骤101中，在接收到上述多个目标运单后，可以通过不同的合包算法确定至少一个合包方案，每个合包方案都包含了所有目标运单，但是目标运单合并成运单包的组合方式并不相同。配送运力对历史运单包的接受行为包括：配送运力曾接受该历史运单包的分配，以及，配送运力曾拒绝该历史运单包的分配。
步骤102，根据每个目标运单包的配送时长，从上述至少一个合包方案中确定目标合包方案。
示例地，该配送时长为以最优路径对该目标运单包进行配送的总时长。可以理解的是，合包后，配送运力以每个运单包为单位对运单进行配送，在此情况下，配送时长越短，则配送运力的效率越高。具体来说，以针对于外卖或快递的配送人员的运单分配为例，外卖或快递运单的分配的基本原则是能够提高配送人员的配送效率，同时降低外卖或快递平台花费的配送成本。其中，提高配送效率在于降低配送时长和减小配送人员被分配的运单的尾单率，降低成本在于使外卖或快递平台在配送人员完成配送任务时支出的邮资等各项费用最小化。基于此，本公开实施例中的合包在整个调度系统中的功能定位也同样是兼顾配送效率的基础上降低配送成本。
示例地，对运单进行合包进而降低成本的原理如下：提高配送人员单次配送行程的配送效率，同时依据运单包顺路度进行邮资折扣。提高配送人员单次配送行程的配送效率意味着配送人员在此次配送行程单位时间收入增加，按顺路度扣减邮资则会使平台在此次配送行程中向配送人员支付的费用减少。两者相结合，具体定价规则可以如下列公式(1)所示：
Figure PCTCN2021093619-appb-000001
其中，T1为以最优路径对该运单包进行组合配送的总时长，T2为每个运单进行单独配送的时长，n为运单包中包含的运单的数量。该公式(1)中，
Figure PCTCN2021093619-appb-000002
即为对于顺路度和邮资折扣之间的关系的数理性描述。需要说明的是，折扣的数值越大，则折扣的力度越小，而折扣的数值越小，则折扣的力度越大。具体地，在运单量和每个运单单独配送的时间一定的情况下，以最优路径对该运单包进行组合配送的总时长越小，则说明上述的多个运单的顺路度越高，进而折扣数值就越小，邮资折扣也就越大。另外，discount_0为预先设定的最小邮资折扣，用于限制该折扣数值不会过小，以保障配送人员的利益。例如，可以将该discount_0设定为0.5(即向配送人员支付预设邮资的50％)，若
Figure PCTCN2021093619-appb-000003
的 计算结果为0.1(即向配送人员支付预设邮资的10％)，则取其中较大的数值，即0.5，进行邮资扣减。
示例地，基于上述的原则，在根据接受行为确定了多个合包方案后，需要对每个合包方案对应的每个运单包的配送时长进行计算，并基于该合包方案的配送时长对这些合包方案进行进一步的筛选，以确定目标合包方案。
步骤103，将该目标合包方案对应的一个或多个目标运单包分配至配送运力。
示例地，在确定了适当的目标合包方案后，可以根据一个地区和/或一个时段内所有配送运力的实际地址和任务量，将目标合包方案中的所有目标运单包分配至相应的配送运力。优选地，配送运力接收到的每个目标运单包，除了包含需要配送的目标运单外，还可以包含针对于每个目标运单包中的所有目标运单的配送计划。该配送计划中可以包含所有目标运单的取件和送件次序。
综上所述，本公开的实施例所提供的技术方案，能够根据配送运力对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，该合包方案用于将上述多个目标运单合并为一个或多个目标运单包，每个目标运单包中的所有目标运单被分配至同一配送运力，该接受度为基于配送运力对历史运单包的接受行为确定的接受度；根据每个目标运单包的配送时长，从上述至少一个合包方案中确定目标合包方案；将该目标合包方案对应的目标运单包分配至配送运力。能够根据配送运力对于已分配的运单包的接受行为和运单包的时间成本确定合包方案，提高运单包与配送运力意愿的契合度，进而提高运单分配效率。
图2是根据图1示出的另一种运单分配方法的流程图，如图2所示，在一实施例中，可以通过预设的接受度预测模型确定配送运力对于历史运单包的接受度，因此，在步骤101之前，该方法还包括接受度预测模型的训练过程，该训练过程可以包括以下步骤。
步骤104，提取多个历史运单包的特征数据和每个历史运单包对应的接受度标记，作为训练数据。
其中，该接受度标记用于表征该配送运力对于历史运单包的接受度。该特征数据包括以下特征中的一个或多个：运单特征、运单包特征和运单区域特征，该运单特征包括运单包中的每个运单的配送过程的配送数据，该运单包特征用于表征运单包中的多个运单之间的关系，该运单区域特征用于表征运单包对应的配送区域的总体配送数据，该接受度标记用于表征运单包是否曾被配送运力接受。
在本实施例中，对接受度进行二元性的划分，即，通过接受度标记(或称标签)来表征配送运力对于所分配的历史运单包的接受和拒绝行为。如此，运单包对应的接受度标记包括：第一标记和第二标记，该第一标记用于表征运单包被配送运力接受，该第二标记用于表征运单包未被配送运力接受。基于此，对配送运力对于运单包的接受行为的预测可以被作为一个分类问题。其过程可以为：通过历史运单包的运单特征、运单包特征和运单区域特征，以及每个历史运单包的接受度标记对预设的分类模型(或称预测模型)进行训练，再通过训练好的分类模型对当前运单包进行预测。在实际的执行过程中，该第一标记可以表示为1，该第二标记可以表示为0。
示例地，上述的运单特征可以包括：运单包中的每个运单的商家出餐时长、约定的交付时长等。上述的运单包特征可以包括：运单包中的各个运单的取餐点之间的距离以及合包后顺路系数等。上述的运单区域特征可以包括：运单所在区域的当前环境特 征和历史统计特征。该当前环境特征可以包括：区域运单负载状态、区域天气等级和区域当前未接单量等。该历史统计特征可以包括本区域在当前时间点之前的预设时间段内的运单的等餐时长、运单被拒绝的概率、配送运力在该区域的取餐路线所覆盖的平均面积等。可以理解的是，除了上述的运单特征、运单包特征和运单区域特征之外，该训练数据还可以为其他的能够尽量完善地反映运单、运单包和配送区域的特征的数据。并且，优选地，训练数据(即历史运单包)的特征种类和待预测的运单包的特征种类相同。
步骤105，通过该训练数据对预设的预测模型进行训练，以生成接受度预测模型。
示例地，需要说明的是，在步骤104和步骤105中，可以收集大量的特征数据对该预测模型进行一次性的训练，或者，可以在系统中定制模型训练任务，以定期更新接受度预测模型。具体地，该模型训练任务可以为，每个预设时长，例如，一天或一周，收集该预设时长内处理的运单包的特征数据和接受度标记并重新对该接受度预测模型进行训练和更新，以保证该接受度预测模型对于该时间和区域的环境变化的敏感度。
图3是根据图2示出的一种制定合包方案的方法的流程图，如图3所示，该步骤101包括步骤1011至步骤1014。
步骤1011，确定上述多个目标运单对应的一个或多个运单包组。
其中，每个运单包组包括对上述多个目标运单进行合并所获得的一个或多个运单包。
示例地，可以通过多种分组规则对上述多个目标运单进行分组，该分组规则包括但不限于：基于配送位置的分组规则、基于送达时间的分组规则以及基于预先设定的聚类模型确定的分组规则等。
步骤1012，针对于每个运单包组，获取该运单包组中的每个运单包的特征数据。
步骤1013，根据上述每个运单包的特征数据以及预设的接受度预测模型，获取上述每个运单包对应的接受度标记。
示例地，该步骤1013可以包括：将上述每个运单包的特征数据作为该接受度预测模型的输入，以获取该接受度预测模型输出的上述每个运单包对应的接受度标记。该接受度预测模型可以为已经在上述步骤104和105中预先训练好的xgboost模型(extreme Gradient Boosting Tree，极端梯度提升树)、SVM模型(Support Vector Machine,支持向量机)或者神经网络模型。以xgboost模型为例，将每个运单包组中的每个运单包的特征数据作为该接受度预测模型的输入，即可以获得该xgboost模型输出的上述每个运单包对应的接受度标记。
步骤1014，根据上述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将该目标运单包组中包含的一个或多个目标运单包作为该合包方案。
示例地，关于本实施例中的“运单包组”和“合包方案”的表述，每个运单包组对应的元素与每个合包方案对应的元素是一致的，均为上述多个目标运单。在不确定运单的组合方式是否符合接受行为条件的情况下，这些被打包的目标运单被称为一个运单包组，在确定运单的组合方式符合接受行为条件的情况下，这些被打包的目标运单被称为一个合包方案。该接受行为条件可以为，一个运单包组内的所有运单包(或者超过一定占比的运单包)均附带上述的第一标记，即，均被配送运力接受，则该运单包组可以被视为一个合包方案。基于此，该步骤1014包括：获取每个运单包组中第一运单包的目标占比，该第一运单包为具备该第一标记的运单包；将该目标占比大于预设比率阈值的运单包组，作为该目标运单包组；将该目标运单包组中包含的一个或多个目标运单包作为该合包方案。
另外，以该接受度预测模型为xgboost模型为例，在步骤1013中，该xgboost模型的实际输出为接受度预测值，该预测值与配送运力接受该运单包的概率正相关。关于模型输出的接受度预测值，可以在通过预设的召回率确定xgboost模型的阈值后，根据该阈值和接受度预测值确定接受度标记。例如，若运单包的接受度预测值大于该阈值，则认为运单包的接受度标记为第一标记；若运单包的接受度预测值小于或等于该阈值，则认为运单包的接受度标记为第二标记。另一方面，可以根据该接受度预测值对xgboost模型进行回溯，在回溯的过程中，可以根据该接受度预测值确定配送运力接受该运单包的概率，再根据该概率对xgboost模型的目标函数进行修正，进而在模型运行的同时提高xgboost模型的预测准确度。
图4是根据图2示出的一种确定合包方案的方法的流程图，如图4所示，该步骤102包括以下步骤。
步骤1021，针对于每个合包方案，确定该合包方案包含的每个目标运单包的配送时长。
其中，该配送时长为将目标运单包中的所有目标运单配送完毕所需的时长。
步骤1022，获取该合包方案包含的各个目标运单包的配送时长的总和，作为该合包方案的总配送耗时。
步骤1023，将具备最短的总配送耗时的合包方案作为该目标合包方案。
示例地，步骤1021可以包括：通过预设的路径分析算法对每个目标运单包对应的配送路径进行分析和规划后，获取每个目标运单包的最优配送路径，该最优配送路径可以为耗时最短的路径或者路程最短的路径。之后，再对通过该最优配送路径配送该目标运单包中的所有目标运单所需的配送时间进行估计，作为该目标运单包的配送时长。基于提高配送效率的基本原则，需要从所有合包方案中确定单均配送时长最短的合包方案。该单均配送时长可以理解为，通过某种合包方案对多个待配送运单进行合包后确定的、多个待配送运单的配送时长的平均值。可以理解的是，在本实施例中，所有合包方案都是针对上述多个目标运单制定的，无论采用哪种合包方案，目标运单的数量是一定的。总配送耗时除以目标运单的数量，就是一个合包方案的单均配送时长。由于目标运单的总数是一定的，可以直接将总配送耗时最短的合包方案认定为该目标合包方案。
图5是根据图2示出的一种分配运单包的方法的流程图，如图5所示，该步骤103包括以下步骤。
步骤1031，根据多个目标运单对应的预设区域内多个配送运力各自的状态信息，从多个配送运力中确定该目标合包方案中的每个目标运单包对应的目标配送运力。
其中，该状态信息包括：上述每个配送运力当前已被分配的配送任务的时间信息和位置信息。
步骤1032，将该目标配送运力对应的目标运单包中包含的所有目标运单分配至该目标配送运力。
示例地，通常来说，处于同一区域的运单被集合在一起进行合包方案的制定和后续的运单分配。具体地，在确定了目标合包方案后，可以根据本区域内每个配送运力的状态信息，确定能够实现目标运单包配送的目标配送运力。例如，在配送运力A已被分配的配送任务与目标运单包B中的配送任务时间和地点上都不存在冲突的情况下，可以将该配送运力A作为该目标运单包B对应的目标配送运力。在上述步骤1031中重复地为目标合包方案中的每个目标运单包确定目标配送运力，直至上述每个运单包都对应有目标配送运力。在此之后，即可在步骤1032中根据目标运单包和目标配送运力的对 应关系进行运单分配。可以理解的是，每个目标运单包作为一个整体进行分配，即，将目标运单包中包含的所有目标运单分配至其对应的目标配送运力。
综上所述，根据本公开的实施例，能够根据配送运力对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，所述合包方案用于将上述多个目标运单合并为一个或多个目标运单包，每个目标运单包中的所有目标运单被分配至同一配送运力，该接受度为基于配送运力对历史运单包的接受行为确定的接受度；根据每个目标运单包的配送时长，从上述至少一个合包方案中确定目标合包方案；将该目标合包方案对应的目标运单包分配至配送运力。能够根据配送运力对于已分配的运单包的接受行为和运单包的时间成本确定合包方案，提高运单包与配送运力意愿的契合度，进而提高运单分配效率。
图6是根据一示例性实施例示出的一种运单分配装置的框图，如图6所示，该装置600包括：
方案制定模块610，被配置成用于根据配送运力对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，该接受度为根据配送运力对历史运单包的接受行为确定的接受度，该合包方案用于将上述多个目标运单合并为一个或多个目标运单包；
方案确定模块620，被配置成用于根据每个目标运单包的配送时长，从上述至少一个合包方案中确定目标合包方案；
运单分配模块630，被配置成用于将该目标合包方案对应的一个或多个目标运单包分配至配送运力。
可选的，该方案制定模块610，被配置成用于：
确定上述多个目标运单对应的一个或多个运单包组，每个运单包组包括对上述多个目标运单进行合并所获得的一个或多个运单包；
针对于每个运单包组，获取该运单包组中的每个运单包的特征数据；
根据所述每个运单包的特征数据以及预设的接受度预测模型，获取上述每个运单包对应的接受度标记；
根据上述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将该目标运单包组中包含的一个或多个目标运单包作为该合包方案。
可选的，所述接受度标记包括第一标记和第二标记，所述第一标记用于表征运单包被配送运力接受，所述第二标记用于表征运单包不被配送运力接受，该方案制定模块610，被配置成用于：
获取每个运单包组中第一运单包的目标占比，该第一运单包为具备该第一标记的运单包；
将该目标占比大于预设比率阈值的运单包组，作为该目标运单包组；
将该目标运单包组中包含的一个或多个目标运单包作为该合包方案。
可选的，图7是根据图6示出的另一种运单分配装置的框图，可以通过预设的接受度预测模型确定配送运力对于运单包的接受度，在一实施例中，可以通过下列的数据提取模块640和模型训练模块650实现接受度预测模型的训练过程，如图7所示，该装置600还包括：
该数据提取模块640，被配置成用于提取多个历史运单包的特征数据和每个历史运单包对应的接受度标记，作为训练数据；
该模型训练模块650，被配置成用于通过该训练数据对预设的预测模型进行训练，以生成接受度预测模型。
可选的，运单包的特征数据以下特征中的一个或多个包括：运单特征、运单包特征和运单区域特征，其中，该运单特征包括运单包中的每个运单的配送过程的配送数据，该运单包特征用于表征运单包中的多个运单之间的关系，该运单区域特征用于表征运单包对应的配送区域的配送数据。
可选的，该方案确定模块620，被配置成用于：
针对于每个合包方案，确定该合包方案包含的每个目标运单包的配送时长，该配送时长为将目标运单包中的所有目标运单配送完毕所需的时长；
获取该合包方案包含的各个目标运单包的配送时长的总和，作为该合包方案的总配送耗时；
将具备最短的总配送耗时的合包方案作为该目标合包方案。
可选的，该运单分配模块630，被配置成用于：
根据所述多个目标运单对应的预设区域内多个配送运力各自的状态信息，从所述多个配送运力中确定该目标合包方案中的每个目标运单包对应的目标配送运力，该状态信息包括：每个配送运力当前已被分配的配送任务的时间信息和位置信息；
将该目标配送运力对应的目标运单包中包含的所有目标运单分配至该目标配送运力。
综上所述，根据本公开的实施例，能够根据配送运力对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，该合包方案用于将上述多个目标运单合并为一个或多个目标运单包，每个目标运单包中的所有目标运单被分配至同一配送运力，该接受度为基于配送运力对历史运单包的接受行为确定的接受度；根据每个目标运单包的配送时长，从上述至少一个合包方案中确定目标合包方案；将该目标合包方案对应的目标运单包分配至配送运力。能够根据配送运力对于已分配的运单包的接受行为和运单包的时间成本确定合包方案，提高运单包与配送运力意愿的契合度，进而提高运单分配效率。
示例地，图8是根据一示例性实施例示出的一种电子设备800的框图。例如，该电子设备800可以被提供为一服务器。参照图8，电子设备800包括处理器801，其数量可以为一个或多个，以及存储器802，其配置为存储可由处理器801执行的计算机程序。存储器802中存储的计算机程序可以包括一个或一个以上的每一个对应于一组指令的模块。此外，处理器801可以被配置为执行该计算机程序，以执行上述图1至图4所示的运单分配方法。
另外，电子设备800还可以包括电源组件803和通信组件804，该电源组件803可以被配置为执行电子设备800的电源管理，该通信组件804可以被配置为实现电子设备800的通信，例如，有线或无线通信。此外，该电子设备800还可以包括输入/输出(I/O)接口805。电子设备800可以操作基于存储在存储器802的操作系统，例如Windows ServerTM，Mac OS XTM，UnixTM，LinuxTM等等。
在另一示例性实施例中，还提供了一种包括程序指令的计算机可读存储介质，该程序指令被处理器执行时实现上述图1至图4所示的运单分配方法的步骤。例如，该计算机可读存储介质可以为上述包括程序指令的存储器802，上述程序指令可由电子设备800的处理器801执行以完成上述图1至图4所示的运单分配方法。
以上结合附图详细描述了本公开的优选实施方式，但是，本公开并不限于上述实施方式中的具体细节，在本公开的技术构思范围内，可以对本公开进行多种简单变型，这些简单变型均属于本公开的保护范围。
另外需要说明的是，在上述具体实施方式中所描述的各个具体技术特征，在不矛盾的情况下，可以通过任何合适的方式进行组合，为了避免不必要的重复，本公开对各种可能的组合方式不再另行说明。

Claims (10)

  1. 一种运单分配方法，所述方法包括：
    根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，
    其中，所述接受度为基于配送资源对历史运单包的接受行为确定的接受度，以及
    所述合包方案用于将所述多个目标运单合并为一个或多个目标运单包；
    根据每个目标运单包的配送时长，从所述至少一个合包方案中确定目标合包方案；
    将所述目标合包方案对应的目标运单包分配至配送资源。
  2. 根据权利要求1所述的方法，其中，所述根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，包括：
    确定所述多个目标运单对应的一个或多个运单包组，每个运单包组包括对所述多个目标运单进行合并所获得的一个或多个运单包；
    针对于每个运单包组，获取该运单包组中的每个运单包的特征数据；
    根据所述每个运单包的特征数据以及预设的接受度预测模型，获取所述每个运单包对应的接受度标记；
    根据所述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案。
  3. 根据权利要求2所述的方法，其中，所述接受度标记包括第一标记和第二标记，所述第一标记用于表征运单包被配送资源接受，所述第二标记用于表征运单包不被配送资源接受，
    所述根据所述每个运单包对应的接受度标记，确定一个或多个目标运单包组，以将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案，包括：
    获取所述每个运单包组中第一运单包的目标占比，所述第一运单包为具备所述第一标记的运单包；
    将所述目标占比大于预设比率阈值的运单包组，作为所述目标运单包组；
    将所述目标运单包组中包含的一个或多个目标运单包作为所述合包方案。
  4. 根据权利要求2所述的方法，其中，所述接受度预测模型采用以下方式得到：
    提取多个历史运单包的特征数据和每个历史运单包对应的接受度标记，作为训练数据；
    通过所述训练数据对预设的预测模型进行训练，以生成所述接受度预测模型。
  5. 根据权利要求2至4中任一项所述的方法，其中，运单包的特征数据包括以下特征中的一个或多个：运单特征、运单包特征和运单区域特征，其中，所述运单特征包括运单包中的每个运单的配送过程的配送数据，所述运单包特征用于表征运单包中的多个运单之间的关系，所述运单区域特征用于表征运单包对应的配送区域的配送数据。
  6. 根据权利要求1所述的方法，其中，所述根据每个目标运单包的配送时长，从所述至少一个合包方案中确定目标合包方案，包括：
    针对于每个合包方案，
    确定该合包方案包含的每个目标运单包的配送时长，所述配送时长为将该目标运单包中的所有目标运单配送完毕所需的时长；
    获取该合包方案包含的各个目标运单包的配送时长的总和，作为该合包方案的总配送耗时；
    将具备最短的总配送耗时的合包方案作为所述目标合包方案。
  7. 根据权利要求1所述的方法，其中，所述将所述目标合包方案对应的目标运单包分配至配送资源，包括：
    根据所述多个目标运单对应的预设区域内多个配送资源各自的状态信息，从所述多个配送资源中确定所述目标合包方案中的每个目标运单包对应的目标配送资源，所述状 态信息包括：每个配送资源当前已被分配的配送任务的时间信息和位置信息；
    将所述目标配送资源对应的目标运单包中包含的所有目标运单分配至所述目标配送资源。
  8. 一种运单分配装置，所述装置包括：
    方案制定模块，被配置成用于根据配送资源对于历史运单包的接受度，为多个目标运单制定至少一个合包方案，其中，所述接受度为基于配送资源对历史运单包的接受行为确定的接受度，所述合包方案用于将所述多个目标运单合并为一个或多个目标运单包；
    方案确定模块，被配置成用于根据每个目标运单包的配送时长，从所述至少一个合包方案中确定目标合包方案；以及，
    运单分配模块，被配置成用于将所述目标合包方案对应的一个或多个目标运单包分配至配送资源。
  9. 一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现权利要求1至7中任一项所述的运单分配方法。
  10. 一种电子设备，包括：
    存储器，其上存储有计算机程序；
    处理器，用于执行所述存储器中的所述计算机程序，以实现权利要求1至7中任一项所述的运单分配方法。
PCT/CN2021/093619 2020-05-13 2021-05-13 一种运单分配方法、装置、存储介质和电子设备 WO2021228198A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010403515.1A CN113673736A (zh) 2020-05-13 2020-05-13 运单分配方法、装置、存储介质和电子设备
CN202010403515.1 2020-05-13

Publications (1)

Publication Number Publication Date
WO2021228198A1 true WO2021228198A1 (zh) 2021-11-18

Family

ID=78525913

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/093619 WO2021228198A1 (zh) 2020-05-13 2021-05-13 一种运单分配方法、装置、存储介质和电子设备

Country Status (2)

Country Link
CN (1) CN113673736A (zh)
WO (1) WO2021228198A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114663169A (zh) * 2022-05-25 2022-06-24 浙江口碑网络技术有限公司 订单数据的处理方法及装置、存储介质、计算机设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107292709A (zh) * 2017-06-14 2017-10-24 北京小度信息科技有限公司 订单处理方法及装置
CN107392412A (zh) * 2017-06-05 2017-11-24 北京小度信息科技有限公司 订单调度方法和装置
CN107748923A (zh) * 2016-08-29 2018-03-02 北京三快在线科技有限公司 订单处理方法、装置及服务器
CN109214551A (zh) * 2018-08-08 2019-01-15 北京三快在线科技有限公司 一种配送调度方法及装置
CN109636213A (zh) * 2018-12-19 2019-04-16 拉扎斯网络科技（上海）有限公司 订单分配、评价方法及装置、电子设备及存储介质
US20190355032A1 (en) * 2014-09-25 2019-11-21 Huawei Technologies Co., Ltd. Order processing method and terminal

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190355032A1 (en) * 2014-09-25 2019-11-21 Huawei Technologies Co., Ltd. Order processing method and terminal
CN107748923A (zh) * 2016-08-29 2018-03-02 北京三快在线科技有限公司 订单处理方法、装置及服务器
CN107392412A (zh) * 2017-06-05 2017-11-24 北京小度信息科技有限公司 订单调度方法和装置
CN107292709A (zh) * 2017-06-14 2017-10-24 北京小度信息科技有限公司 订单处理方法及装置
CN109214551A (zh) * 2018-08-08 2019-01-15 北京三快在线科技有限公司 一种配送调度方法及装置
CN109636213A (zh) * 2018-12-19 2019-04-16 拉扎斯网络科技（上海）有限公司 订单分配、评价方法及装置、电子设备及存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114663169A (zh) * 2022-05-25 2022-06-24 浙江口碑网络技术有限公司 订单数据的处理方法及装置、存储介质、计算机设备

Also Published As

Publication number Publication date
CN113673736A (zh) 2021-11-19

Similar Documents

Publication Publication Date Title
US11238378B2 (en) Method and system for booking transportation services
AU2017255282B2 (en) System and method for determining routes of transportation service
JP6655939B2 (ja) 輸送サービス予約方法、輸送サービス予約装置、及び輸送サービス予約プログラム
WO2018095066A1 (zh) 任务分组方法、装置、电子设备及计算机存储介质
JP2017165509A (ja) 運送管理システム
US20090276267A1 (en) Apparatus and method for handling weight data related to transportation
US10740682B2 (en) Sensor based truth maintenance
CN112529487B (zh) 车辆调度方法、装置以及存储介质
CN112183852A (zh) 物流配送路线筛选及运费核算方法、系统、终端及介质
CN111340318B (zh) 一种车辆动态调度方法、装置及终端设备
CN111695842B (zh) 配送方案确定方法、装置、电子设备及计算机存储介质
WO2021228198A1 (zh) 一种运单分配方法、装置、存储介质和电子设备
CN111626482A (zh) 航空货运舱位分配方法及系统
Schönberger Scheduling constraints in dial-a-ride problems with transfers: a metaheuristic approach incorporating a cross-route scheduling procedure with postponement opportunities
CN109583634A (zh) 一种基于现代投资组合理论的外卖配送路径选择方法
CN114693004B (zh) 物流优化方法和装置
CN116777171A (zh) 一种面向网约车的点对点拼车动态调度方法及装置
CN116843121A (zh) 基于等级匹配度的移动群智感知任务分配方法及管理系统
CN111210074A (zh) 仓库中的订单处理方法、装置、介质、电子设备与系统
CN112819394A (zh) 运单处理方法、装置、计算机可读存储介质及电子设备
Fragkos et al. Supply planning for shelters and emergency management crews
CN111833595B (zh) 共享汽车辅助车辆配置方法、电子设备及存储介质
WO2021121348A1 (en) Cumulative surged ride value calculation on a ridesharing platform
CN117314132B (zh) 一种基于大数据的调度方法及系统
US11782662B1 (en) Method and system for asset routing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21804683

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21804683

Country of ref document: EP

Kind code of ref document: A1