CN109919532A - Logistics node determination method and device - Google Patents

Abstract

The embodiment of the application provides a method and a device for determining logistics nodes, relates to the technical field of logistics, and comprises the following steps: determining estimated logistics list quantity from a plurality of primary nodes of a delivery place to secondary nodes of a receiving place; determining logistics cost parameters from a plurality of primary nodes of the delivery places to a plurality of secondary nodes of the candidate delivery places; according to the estimated logistics list quantity and the logistics cost parameters, the node distribution model is utilized to determine the target delivery place secondary node and the delivery place primary node corresponding to the target delivery place secondary node, the purpose of adding the assembly straight line on the basis of the original straight line is achieved, the problems that the number of the existing straight lines is small, and the express aggregation time in the starting point of the existing straight line is long are solved, and the logistics efficiency is improved and the logistics cost is reduced.

Description

A kind of logistics node determines method and device

Technical field

This application involves logistics technology, determine method and device more particularly to a kind of logistics node.

Background technique

With the development of delivery industry, express delivery amount is also increasing.But the quick despatch amount rapid growth the case where Under, the quantity of logistics provider transshipment center is not significantly increased；Meanwhile being limited by the field, the processing capacity of transshipment center begins It can not break through eventually.

In general, the logistics provider of e-commerce cooperation, with place of acceptance first nodes dimension, (site is sent to place of acceptance first nodes Single amount) average waybill amount in statistics each site the past period, using this waybill amount that is averaged as correspondence net Daily single amount of point, if daily single amount of site is more than preset threshold, then it is assumed that the courier packages of the site can collect The straight hair route of the site with corresponding place of acceptance first nodes is opened in middle dispatching at this time, if daily single amount of site is not above The courier packages of site are sent to place of acceptance first nodes by transfer station then according to old process by preset threshold.

But it is more demanding to the volume of goods transported of site by opening for this logistics mode progress straight hair route, also, by It is limited in the poly- goods ability in single site itself by factors such as regions, carries out the straight of a certain site according to above-mentioned logistics mode Line clear is sent out, need to wait until that the package quantity of the site reaches a biggish preset threshold, this results in straight hair routes Service time is longer, therefore within a certain period of time, so that the dot number for meeting volume of freight is less, so as to cause the efficiency of logistics It is lower.

Summary of the invention

The technical problem to be solved in the embodiments of the present application is that providing a kind of logistics node determines method, to solve existing object It is more demanding to the volume of goods transported of site that stream mode carries out opening for straight hair route, so that the dot number for meeting volume of freight is less, So as to cause the lower problem of the efficiency of logistics.

Correspondingly, the embodiment of the present application also provides a kind of logistics node determining devices, to guarantee the reality of the above method Existing and application.

To solve the above-mentioned problems, the embodiment of the present application discloses a kind of logistics node and determines method, comprising:

Determine that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node；

Determine the multiple place of departure first nodes to multiple candidate place of departure two-level nodes logistics cost parameter；

Logistics list amount and logistics cost parameter are estimated according to described, using node distribution model, determines target place of departure Two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

The embodiment of the present application discloses a kind of logistics node determining device, comprising:

First determining module, for determining that multiple place of departure first nodes estimate logistics list to place of acceptance two-level node Amount；

Second determining module, for determining the multiple place of departure first nodes to multiple candidate place of departure two-level nodes Logistics cost parameter；

Node distribution module distributes mould using node for estimating logistics list amount and logistics cost parameter according to Type determines target place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

Correspondingly, the embodiment of the present application also discloses a kind of device characterized by comprising

One or more processors；With

One or more machine readable medias of instruction are stored thereon with, are executed when by one or more of processors When, so that described device executes a kind of logistics node and determines method.

Correspondingly, the embodiment of the present application also discloses one or more machine readable medias, be stored thereon with instruction, when by When one or more processors execute, so that device executes a kind of logistics node and determines method.

The embodiment of the present application includes the following advantages:

A kind of logistics node provided by the embodiments of the present application determines method and device, passes through the multiple place of departure level-one sections of determination Point estimates logistics list amount to place of acceptance two-level node；Determine multiple place of departure first nodes to multiple candidate place of departure second level sections The logistics cost parameter of point；According to logistics list amount and logistics cost parameter is estimated, using node distribution model, determine that target is sent out Goods ground two-level node and the corresponding place of departure first nodes of target place of departure two-level node, have reached in original straight hair route On the basis of increase the purpose for spelling goods straight hair route, the package of the multiple sender nodes of rapid aggregation a to place of departure can be passed through Two-level node carries out the open-minded of additional straight hair route, and the time needed for poly- goods to a place of departure two-level node is shorter, solves The long problem of logistics package assemble index, increases the quantity of straight hair route, improves object in the starting point of existing straight hair route It flows efficiency, and due to combining the determination process estimated logistics list amount and logistics cost parameter and participate in logistics node, allows package The path of poly- goods is short as far as possible, so that logistics node is determined to be optimized according to logistics cost parameter, having is reduced The beneficial effect of logistics cost.

Detailed description of the invention

Fig. 1 is the processing schematic that one of the embodiment of the present application determines method based on logistics node；

Fig. 2 is the step flow chart that one of the embodiment of the present application logistics node determines method；

Fig. 3 is the specific steps flow chart that one of the embodiment of the present application logistics node determines method；

Fig. 3 A is the acceptance criterion function schematic diagram of one of the embodiment of the present application simulated annealing；

Fig. 4 is the logistics transportation method and step process that one of the embodiment of the present application determines method based on logistics node Figure；

Fig. 5 is the structure chart of one of the embodiment of the present application logistics node determining device；

Fig. 6 is the concrete structure diagram of one of the embodiment of the present application logistics node determining device；

Fig. 7 is the hardware structural diagram of one of the embodiment of the present application device.

Specific embodiment

In order to make the above objects, features, and advantages of the present application more apparent, with reference to the accompanying drawing and it is specific real Applying mode, the present application will be further described in detail.

A kind of logistics node provided by the present application determines that term common in method has:

Place of acceptance first nodes: the last one transshipment center of courier packages during logistics transportation, courier packages arrive Up to after place of acceptance first nodes, can be sent with charge free by courier in client's hand.

Place of acceptance two-level node: being divided into multiple consolidating the load regions by routing rule, and the fortune largely wrapped up can be supported by having The place turned, and position is in the place of acceptance first nodes at the center in consolidating the load region as far as possible, which can be used as place of acceptance two Grade node.

Place of departure first nodes: first transshipment center of courier packages during logistics transportation, for collecting export Courier packages.

Place of departure two-level node: special aggregation package spell the site of goods straight hair, for collecting multiple corresponding places of departure The package that first nodes are sent opens place of departure two-level node to receipts when the package quantity received meets straight hair requirement Goods two-level node straight hair route.

Currently, the entire transmission flow of an express delivery is: addressee-arrival place of departure first nodes-arrival place of departure Relay centre-arrival place of acceptance two-level node-arrival place of acceptance first nodes-courier send with charge free-signs for, it is seen that in mesh Before, the dispatching of express delivery needs to connect place of departure first nodes and place of acceptance two-level node by relay centre, therefore, in quick despatch In the case where measuring rapid growth, it is limited by the field, the processing capacity of relay centre can not be broken through always.

The concept of express delivery straight hair, which refers to, is directly sent to place of acceptance two-level node, straight hair mistake from place of departure two-level node for express delivery There is no relay centre among journey, transmitting efficiency can be greatly improved in this way, and reduce transportation cost.In the embodiment of the present application In the specific implementation, can will spell goods straight hair and resolve into three parts: first part is this application provides the concept for spelling goods straight hair Single amount is estimated, and second part is the poly- goods in rear end, and Part III is the poly- goods in front end.

Firstly, the poly- goods in rear end be under the premise of not influencing timeliness, will be similar in shipping address in courier packages destination Place of acceptance first nodes carry out region merging technique, and choose one of place of acceptance first nodes as place of acceptance two-level node, The meaning of place of acceptance two-level node is the terminal as straight hair route, and increases single amount number of straight hair route.

Secondly, it is by server according to each place of departure first nodes going through to place of acceptance two-level node route that single amount, which is estimated, History logistics list amount estimates each place of departure first nodes to place of acceptance two-level node route and estimates logistics list amount.

Again, the poly- goods in front end is the site work chosen in each place of departure first nodes for corresponding place of acceptance two-level node It can make each hair by matching corresponding place of departure two-level node for each place of departure first nodes for place of departure two-level node Goods the express delivery drawn over to one's side of first nodes be gathered in corresponding place of departure two-level node, when the received package number of place of departure two-level node It when reaching the package number of straight hair requirement, opens using the place of departure two-level node as starting point, the corresponding receipts of place of departure two-level node Goods two-level node be terminal straight hair route.By the poly- goods of the poly- goods in front end and rear end, increase on the basis of conventional straight hair route Additional spelling goods straight hair route is added, has improved express delivery straight hair efficiency, save cost.

It should be noted that the place of acceptance two-level node and place of departure two-level node in the embodiment of the present application need to possess energy It enough supports the place for the operating largely wrapped up and present position is in its corresponding regional center as far as possible.

Referring to Fig.1, the processing schematic that one of the embodiment of the present application determines method based on logistics node is shown.

In the logistics transportation method based on place of departure two-level node in the specific implementation, logistics node provided by the present application is true Determine method just will carry out once planning based on the whole network for spelling goods straight hair route every setting time interval, wherein setting time interval For example several moons, the embodiment of the present application do not limit it.Referring to Fig.1, module is estimated by single amount first, when extracting default Interior place of departure first nodes estimate each delivery according to history logistics list amount to the history logistics list amount of each place of acceptance first nodes Ground first nodes estimated sub- logistics list amount by following several months of each place of acceptance first nodes.

Furthermore it is possible to each place of acceptance first nodes are divided into multiple consolidating the load regions by the first preset rules, by shipping area The place for the operating largely wrapped up can be supported in domain and present position is in the place of acceptance one of its corresponding regional center as far as possible Grade node is as candidate place of acceptance two-level node, for example, having place of acceptance first nodes 1a, place of acceptance level-one in Fig. 1 example Tri- level-ones of node 1b, place of acceptance first nodes 1c are received node, choose the field for waiting the operating that can support largely to wrap up at this time Ground and present position are in the place of acceptance first nodes 1c of its corresponding regional center as candidate place of acceptance two-level node as far as possible 1, candidate place of acceptance two-level node is used for other place of acceptance level-one sections being dispatched into the courier packages received in corresponding region Point, it should be noted that the division in consolidating the load region is fulfiled ahead of schedule by the national logistics pool of logistics company.Specific such as Fig. 1, Place of acceptance first nodes 1a, place of acceptance first nodes 1b, two-level node 1 is in same consolidating the load region with receiving candidate goods.

It counts multiple place of departure first nodes and estimates logistics list amount, such as candidate place of acceptance two to place of acceptance two-level node The logistics list amount of estimating of grade node 1 is that place of departure first nodes 1a, place of departure first nodes 1b and place of departure two-level node 1 are sent To the summation for estimating sub- logistics list amount in 1 place consolidating the load region of place of acceptance two-level node, wherein place of departure two-level node 1 is sent Sub- logistics list amount is estimated to 1 place consolidating the load region of place of acceptance two-level node, refers to that place of departure two-level node 1 is sent out as one Goods first nodes when be sent to corresponding consolidating the load region estimate sub- logistics list amount.

Further, it can also be directed to each consolidating the load region, selected from each place of departure first nodes by the second preset rules Candidate place of departure two-level node in figure, has place of departure first nodes 1a, place of departure first nodes 1b, place of departure first nodes Tri- level-one sender nodes of 1c, selection at this time, which is waited, can support the place for the operating largely wrapped up and present position is in it as far as possible The place of departure first nodes 1c of corresponding regional center is as candidate place of departure two-level node 1, candidate 1 needle of place of departure two-level node To the corresponding consolidating the load region of candidate place of acceptance two-level node 1, and it can receive place of departure first nodes 1a and place of departure level-one section The spelling package of point 1b is wrapped up in.

Finally, be based on node distribution model by node distribution module, select a candidate place of departure two-level node as Target place of departure two-level node, and determine the corresponding place of departure first nodes of target place of departure two-level node, i.e. place of departure level-one After node 1a, place of departure first nodes 1b receive the distribution information of the candidate place of departure two-level node 1 of server transmission, determine Candidate place of departure two-level node 1 is target place of departure two-level node 1, will pull and receives the express delivery for being sent to corresponding consolidating the load region aggregation It is sent to target place of acceptance two-level node 1 to the target place of departure two-level node 1 of distribution, and by target place of departure two-level node 1, The courier packages received are sent to corresponding place of acceptance first nodes according to shipping address by target place of acceptance two-level node 1 1a, place of acceptance first nodes 1b, is dispatched into customer's hand finally by old process.

Referring to Fig. 2, a kind of logistics node for showing the application determines the step flow chart of method, can specifically include as Lower step:

Step 201, determine that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node.

In the embodiment of the present application, all place of departure first nodes are stored in the database of logistics system in preset time The interior history logistics list amount data to each place of acceptance first nodes, single amount is estimated can be by these history logistics list amount data One or more, estimate place of departure first nodes in the daily logistics list amount of following a period of time.

In the embodiment of the present application, count place of acceptance two-level node estimates logistics list amount, can be used as subsequent node point The function of node distribution model is improved in input when with model solution.

It should be noted that for the place of acceptance two-level node of rear end, for each place of acceptance two-level node, statistics delivery Ground first nodes estimate logistics list amount to place of acceptance two-level node, it can be determined that this estimates whether logistics list amount reaches straight hair list The requirement of amount.Subsequent step can be executed to the place of acceptance two-level node for estimating logistics list amount and reaching the requirement of straight hair list amount, to not having The place of acceptance two-level node for reaching the requirement of straight hair list amount does not execute subsequent step, saves calculation amount, saves overhead.

Such as: the place of acceptance two-level node for place of departure first nodes is Beijing, Harbin, Tianjin, Huhehaote etc. Single amount in city, can be merged into Pekinese and singly measures, input when solving as subsequent node distribution model.

Specifically, estimating place of departure first nodes estimating sub- logistics list amount and can pass through system to each place of acceptance first nodes Count the average value of history logistics list amount in preset time；Pass through time series models；It is obtained by the methods of regression model, this three What kind of method obtained, which estimate sub- logistics list amount, covers the nearly all factor for influencing the variation of logistics list amount, it is possible to this three The sub- logistics list amount of estimating that kind method obtains is weighted and averaged, and is obtained most accurately estimating sub- logistics list amount, is needed to illustrate It is, for each place of acceptance two-level node, to require to carry out above-mentioned logistics list amount to estimate operation.

Step 202, determine the multiple place of departure first nodes to multiple candidate place of departure two-level nodes logistics cost Parameter.

In the embodiment of the present application, logistics cost of multiple place of departure first nodes to multiple candidate place of departure two-level nodes Parameter refers to each place of departure first nodes to the distance between each place of departure two-level node in corresponding consolidating the load region, the logistics Cost parameter data storage can extract calling in the database of logistics system by relevant interface.Wherein, logistics at This parameter may include distance, can also be including logistics cost etc., and the embodiment of the present application does not limit it.For example, logistics Cost parameter can be that the distance of place of departure first nodes A to candidate place of departure two-level node a are 5.3 kilometers, place of departure level-one The distance of node B to candidate place of departure two-level node b are 1.7 kilometers.

Step 203, it logistics list amount and logistics cost parameter is estimated determines mesh according to described using node distribution model Mark place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

In the embodiment of the present application, by preset node distribution model, each place of departure first nodes are received to correspondence Ground two-level node estimates logistics list amount and each place of departure first nodes to each place of departure second level in corresponding consolidating the load region The distance between node, by genetic algorithm, under conditions of meeting straight hair list amount range, with the smallest each place of departure level-one section The sum of the distance of point to corresponding place of departure two-level node is target, and it is preferred to distribute at least one for each place of departure first nodes Place of departure two-level node, place of departure first nodes can choose any one from multiple preferred place of departure two-level nodes, or Genetic algorithm can be helped to jump out local optimum by Simulated Anneal Algorithm Optimize genetic algorithm to the selection strategy of parent's body Solution.

For example, a collection of express delivery is sent to Beijing, Tianjin, Kazakhstan from Hangzhou, Shaoxing, Ningbo, Wuxi, Xi'an, Xianyang, Baoji correspondence That shore, Shijiazhuang.But it is directed to Hangzhou-Beijing, Shaoxing-Tianjin, Ningbo-Harbin, Wuxi-Shijiazhuang route, Hangzhoupro State, Xi'an history list amount be greater than the first history list amount threshold value, illustrate that Hangzhou, Xi'an meet place scale requirements, at this time can be with In the Hangzhou for belonging to the same area, Shaoxing, Ningbo, Ningbo is chosen as place of departure two-level node in Wuxi, belong to the same area Xi'an, Xianyang, select Xi'an as place of departure two-level node in Baoji.And Shaoxing, Ningbo, Wuxi to Hangzhou, Xianyang, Baoji Division to Xi'an keeps logistics cost minimum.So Shaoxing, Ningbo, Wuxi are sent to Beijing, Tianjin, Harbin, Shijiazhuang Logistics package is sent to Hangzhou, and the logistics that Xianyang, Baoji are sent to Beijing, Tianjin, Harbin, Shijiazhuang, which is wrapped up, is sent to west Peace.Simultaneously can in the Beijing for belonging to the same area, Tianjin, Harbin, choose Beijing as place of acceptance second level section in Shijiazhuang Hangzhou-Beijing, Xi'an-Pekinese's straight hair route are opened after consolidating the load to Hangzhou in point, Shaoxing, Ningbo, Wuxi, and express delivery reaches Behind Beijing, then by Beijing Tianjin, Harbin, Shijiazhuang are dispatched into according to the place of acceptance of express delivery.Thus in original Hangzhou-Beijing Ningbo-Pekinese's straight hair route is increased on the basis of one straight hair route, improves logistic efficiency.

In conclusion a kind of logistics node provided by the embodiments of the present application determines method, multiple place of departure level-ones are determined excessively Node estimates logistics list amount to place of acceptance two-level node；Determine multiple place of departure first nodes to multiple candidate place of departure second levels The logistics cost parameter of node；Target is determined using node distribution model according to logistics list amount and logistics cost parameter is estimated Place of departure two-level node and the corresponding place of departure first nodes of target place of departure two-level node, have reached in original straight hair route On the basis of increase the purpose for spelling goods straight hair route, can be delivered by the packages of the multiple sender nodes of rapid aggregation to one Ground two-level node carries out the open-minded of additional straight hair route, and the time needed for poly- goods to a place of departure two-level node is shorter, solution The long problem of logistics package assemble index, increases the quantity of straight hair route, Ke Yiti in the starting point of existing straight hair route of having determined High logistic efficiency, and due to combining the determination process estimated logistics list amount and logistics cost parameter and participate in logistics node, it allows The path for wrapping up poly- goods is short as far as possible, so that logistics node is determined to be optimized according to logistics cost parameter, has Reduce the beneficial effect of logistics cost.

Referring to Fig. 3, a kind of logistics node for showing the application determines the specific steps flow chart of method, specifically can wrap Include following steps:

Step 301, candidate place of acceptance two-level node is selected from each place of acceptance first nodes according to the first preset rules.

In the embodiment of the present application, the first preset rules can be routing rule, and routing rule refers to router according to road By the information in table, an optimal path is selected, data forwarding is gone out.Routing rule can be applied to logistics field, i.e., According to the characteristic of the place of acceptance first nodes stored in logistics service device, the place of acceptance that will be had a good transport service or neighbor distance is close First nodes are divided into corresponding consolidating the load region, with meet between the place of acceptance first nodes in consolidating the load region traffic convenience and Neighbor distance is close, and the characteristic of place of acceptance first nodes includes the scale of place of acceptance first nodes, receives apart from other The distance of ground first nodes is closely remote, traffic convenience degree of present position etc., such as national consolidating the load region can according to routing rule To be divided into North China region, South China Regional, several overall areas such as northwest region, each overall area can also be further subdivided into several A zonule, to advanced optimize the function of rear end consolidating the load.

Further, after having divided consolidating the load region, according to routing rule, choosing in consolidating the load region can support largely to wrap up Operating place and present position be in its corresponding regional center as far as possible and history receiving note amount be greater than the second history list The place of acceptance first nodes of threshold value are measured as place of acceptance two-level node, after place of acceptance two-level node receives express delivery, then by receiving Goods two-level node send corresponding place of acceptance first nodes according to place of acceptance for express delivery.Improve the efficiency of express transportation. It is greater than in the place of acceptance first nodes of second threshold it should be noted that can choose history receiving note amount, near field The place of acceptance first nodes of the heart are as place of acceptance two-level node.

For example, can be by the express delivery of the place of acceptances first nodes such as Beijing, Harbin, Tianjin, Huhehaote, unified elder generation's straight hair Final place of acceptance first nodes are dealt into Beijing, then by Beijing, Beijing is used as Harbin, Tianjin, Huhehaote affiliated area Candidate place of acceptance two-level node.

Optionally, step 301 can also include sub-step 3011 and sub-step 3012.

Each place of acceptance first nodes are divided into multiple consolidating the load regions by routing rule by sub-step 3011.

In the embodiment of the present application, routing rule refers to router according to the information in routing table, and selection one is optimal Data forwarding is gone out in path.Routing rule can be applied to logistics field, i.e., according to the place of acceptance stored in logistics service device The place of acceptance first nodes having a good transport service or neighbor distance is close are divided into corresponding consolidating the load region by the characteristic of first nodes In, to meet traffic convenience between each place of acceptance first nodes in consolidating the load region and neighbor distance is close, place of acceptance level-one section The characteristic of point includes the scale of place of acceptance first nodes, and the distance apart from other place of acceptance first nodes is closely remote, institute Locate the traffic convenience degree etc. of position, such as national consolidating the load region can be divided into North China region, south China area according to routing rule Domain, several overall areas such as northwest region, each overall area can also be further subdivided into several zonules, after advanced optimizing Hold the function of consolidating the load.

Sub-step 3012 selects a history logistics list amount from each place of acceptance first nodes in each consolidating the load region Greater than the place of acceptance first nodes of the first preset threshold as the candidate place of acceptance two-level node for being directed to the consolidating the load region.

Further, after having divided consolidating the load region, according to routing rule, choosing in consolidating the load region can support largely to wrap up Operating place and present position be in the place of acceptance first nodes of its corresponding regional center as far as possible as candidate place of acceptance Two-level node, after target place of acceptance two-level node receives express delivery, then by place of acceptance two-level node by express delivery according to place of acceptance It is sent to corresponding place of acceptance first nodes.Improve the efficiency of express transportation.

For example, can be Beijing, Harbin, Tianjin, the isocentric express delivery in Huhehaote, unification by place of acceptance first nodes First straight hair is dealt into final purpose center to Beijing, then by Beijing, and Beijing is used as Harbin, Tianjin, Huhehaote affiliated area Place of acceptance two-level node.

Step 302, determine that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node.

The step is specifically referred to above-mentioned steps 201, and details are not described herein again.

Optionally, step 302 can also include sub-step 3021, sub-step 3022 and sub-step 3023.

Sub-step 3021, obtain the place of departure first nodes to each place of acceptance first nodes history logistics list amount.

In the embodiment of the present application, all place of departure first nodes are stored in the database of logistics system in preset time The interior history logistics list amount data to each place of acceptance first nodes can obtain the history logistics list by calling the corresponding interface Measure data.

Sub-step 3022 estimates the place of departure first nodes to each place of acceptance level-one according to the history logistics list amount Node estimates sub- logistics list amount.

In this step, it is to estimate place of departure first nodes not by these history logistics list amount data that single amount, which is estimated, The daily logistics list amount for carrying out a period of time can filter out the place of departure one for meeting straight hair list amount by the single amount data estimated Grade node and delivers to the place of departure first nodes for being unsatisfactory for straight hair list amount to the straight hair route of place of acceptance two-level node The selection of ground two-level node, with meet the corresponding each place of departure first nodes of place of departure two-level node estimate sub- logistics list amount it With the requirement that can reach straight hair list amount.

Optionally, step 3022 can also include sub-step 30221, sub-step 30222 and sub-step 30223.

The place of departure first nodes are arrived the history of each place of acceptance first nodes by sub-step 30221 within a preset time Logistics list amount and the corresponding factor parameter of history logistics list amount input regression model, obtain the place of departure first nodes to each receipts Goods first nodes estimate sub- logistics list amount, the factor parameter include: one of inventory information, month, tendency or History logistics a variety of, that the regression model passes through the place of departure first nodes in preset time to each place of acceptance first nodes It is single to measure factor parameter training acquisition corresponding with history logistics list amount.

In practical applications, the history logistics list amount data of express delivery site are influenced by factors, such as: in different seasons Section, customer can be influenced by season and buy different products, such as the certain sites of summer possibility air-conditioning express delivery handling capacity It is more compared with other months, and winter other sites cotton-wadded quilt express delivery handling capacity express delivery site history list amount data compared with it He is more month, or the influence of " double 11 shopping sections " and " Black Friday shopping section ", the meeting of express delivery site are received in November Generate large batch of order.Therefore the history logistics list amount data of express delivery site are by factors such as inventory information, month, tendencys It is affected.

In the embodiment of the present application, it can be preset with regression model, regression model is based on algorithm with regress analysis method, by default The history logistics list amount and history logistics list amount of the place of departure first nodes to each place of acceptance first nodes are corresponding in time Factor training obtains, and regression analysis is the calculation method for studying a variable about the specific dependence of another (a little) variable And theory.Specifically from one group of sample data, the relationship between variable is determined, and to the credible of these relational expressions Degree carries out various statistical checks, and the influence which variable is found out from all multivariables for influencing a certain particular variables is significant, which It is not significant a bit.Using required relational expression, another particular variables is predicted or controlled according to the value of one or several variables Value, and provide it is this prediction or control levels of precision.It should be noted that regression model used herein can make This is not construed as limiting with analysis methods, the application such as linear regression, vector regression, promotion regression trees.

So the foundation of regression model is to arrive each place of acceptance level-one section within a preset time based on each place of departure first nodes History logistics list amount and the impact factor training of point obtain the relationship between history logistics list amount and impact factor, the number The impact factor learned in relational expression is endowed default weight, such as weight > tendency weight > month weight of inventory information. By the relationship, estimate to obtain estimate sub- logistics list amount of the place of departure first nodes to each place of acceptance first nodes, benefit It can be estimated out with impact factor and more accurately estimate sub- logistics list amount, so that entire place of departure two-level node selection method considers Seasonality, the factors such as tendency improve the accuracy of place of departure two-level node selection.

The place of departure first nodes are arrived the history of each place of acceptance first nodes by sub-step 30222 within a preset time Logistics list amount input time series model, obtains the place of departure first nodes to each place of acceptance first nodes and estimates sub- logistics Dan Liang, the time series models arrive going through for each place of acceptance first nodes by the place of departure first nodes within a preset time History logistics list amount and the corresponding date training of history logistics list amount obtain.

In production and scientific research, some or one group of variable x (t) are observed and measured, it will be at a series of moment T1, t2 ..., tn (t is independent variable) are arranged according to chronological order, and are used for the mathematic(al) representation of explanatory variable and correlation, Obtained discrete digital composition sequence set, we term it time serieses.

In the embodiment of the present application, based on the method for time series, history logistics list amount is excavated by time series models The tendency and periodicity of variation, the more accurately single amount that can provide are estimated, and the type of time series models includes but is not limited to Arma model, arch model etc..

Firstly, calculating route nearly 30 days daily single amounts (n1, n2 ..., n30)；Then this 30 days single amounts are input to the time Series model, what last model exported to obtain route estimates sub- logistics list amount.

For example, to predict this place of departure first nodes of Hangzhou Yuhang District to this place of acceptance level-one of Beijing transshipment center Daily single amount of node calculates the route nearly 30 days daily single amounts first and such as (9013,8563 ..., 8921) is input to the time In series model, after training, the sub- logistics list amount of output estimation is 8792.

The place of departure first nodes are arrived the history of each place of acceptance first nodes by sub-step 30223 within a preset time The average value of logistics list amount estimates sub- logistics list amount as the place of departure first nodes to each place of acceptance first nodes.

The place of departure first nodes are arrived the history of each place of acceptance first nodes by sub-step 30224 within a preset time Logistics list amount and the corresponding factor parameter of history logistics list amount input regression model, obtain the place of departure first nodes to each receipts Goods first nodes first estimate sub- logistics list amount.The place of departure first nodes are arrived into each place of acceptance one within a preset time The history logistics list amount input time series model of grade node, obtains the place of departure first nodes to each place of acceptance first nodes Second estimate sub- logistics list amount.The place of departure first nodes are arrived to the history of each place of acceptance first nodes within a preset time The average value of logistics list amount estimates sub- logistics list amount as the third of the place of departure first nodes to each place of acceptance first nodes. Sub- logistics list amount is estimated by first, second estimates sub- logistics list amount and third is estimated sub- logistics list amount and is weighted with default weight It is average, the place of departure first nodes, which are obtained, to each place of acceptance first nodes estimates sub- logistics list amount.

In this step, above-mentioned sub-step 30221, sub-step 30222, sub-step 30223 calculate separately out first and estimate Sub- logistics list amount, second estimate sub- logistics list amount and third estimates sub- logistics list amount, and in these three results, first estimates sub- logistics Single amount is relatively more accurate, it is possible to be weighted and averaged to these three results, it is higher to estimate sub- logistics list amount imparting for first Weight, the result after obtained weighted average be cover most of impact factor and time series factor estimate sub- object The single amount of stream, further improves the accuracy for estimating sub- logistics list amount, and then improves the accurate of place of departure two-level node selection Property.

For example, first estimates that sub- logistics list amount is A, second to estimate sub- logistics list amount be B, third estimates sub- logistics list amount and is C, then it is final to estimate sub- logistics list amount=A × 0.5+B × 0.25+C × 0.25.

Sub-step 3023 counts each place of departure according to the corresponding place of acceptance first nodes of each candidate's place of acceptance two-level node First nodes estimate logistics list amount to each candidate place of acceptance two-level node.

In the embodiment of the present application, the place of acceptance two-level node chosen for step 301, statistics place of acceptance two-level node Logistics list amount is estimated, input when subsequent node distribution model solves is can be used as, improves the function of node distribution model.It needs Illustrating, the quantity for estimating logistics list amount needs to meet the requirement of straight hair list amount, specially in the division in consolidating the load region, with The corresponding place of departure first nodes of place of acceptance first nodes estimates the sum of sub- logistics list amount and to reach straight hair list in consolidating the load region The division in consolidating the load region is carried out for the purpose of amount.

Step 303, candidate place of departure two-level node is selected from each place of departure first nodes according to the second preset rules.

In this application, the net for being not up to the requirement of straight hair list amount for estimating sub- logistics list amount in place of departure first nodes Point can distribute place of departure two-level node for it and carry out poly- goods, and place of departure two-level node mainly undertakes the function of poly- goods point, delivery Ground first nodes monopolize the express delivery collected, and after being sent to the place of departure two-level node of distribution, place of departure two-level node is responsible for poly- goods Straight hair.Place of departure two-level node needs to have several features: 1, having the place that can support the operating largely wrapped up.2, present position It is in center as far as possible.

Optionally, step 303 can also include sub-step 3031 and sub-step 3032.

Sub-step 3031, according to the longitude and latitude of place of departure first nodes present position, by the place of departure level-one section Point is polymerized to multiple site clusters according to clustering algorithm.

Selection Strategy of the application to candidate place of departure two-level node are as follows:

Firstly, clustering to all place of departure first nodes according to longitude and latitude, several site clusters are acquired.Here using based on close The mode for spending cluster, such as DBSCAN (clustering algorithm, English: Density-Based Spatial Clustering of Applications with Noise) etc., DBSCAN is a more representational density-based algorithms.With draw Point different with hierarchy clustering method, cluster is defined as the maximum set for the point that density is connected by it, can be with enough high density Region division be cluster, and the cluster of arbitrary shape can be found in the spatial database of noise.

Sub-step 3032 will estimate sub- logistics list amount greater than the second preset threshold in the site cluster, and, apart from site cluster The nearest place of departure first nodes in center, are determined as candidate place of departure two-level node.

In conjunction with the Selection Strategy of above-mentioned sub-step 3031, further, sub- logistics is estimated according to place of departure first nodes Single amount filters out big site, will be close to the big site at cluster center, is set as place of departure two-level node.

In this step, according to the estimation results of sub-step 3022, sub- logistics list amount will be estimated in place of departure first nodes It is big site greater than the screening of the site of preset threshold, and big site needs the place that can support the operating largely wrapped up, most Afterwards by the big site in the big site for the condition that meets close to cluster center, it is set as place of departure two-level node, place of departure two-level node Advantage be to can satisfy the transhipment of bulky goods, and the traffic convenience with place of departure first nodes around.

The selection of place of departure two-level node improves the poly- goods ability in front end of express delivery straight hair logistics, in conventional straight hair route On the basis of increase spell goods straight hair route, reduce the cost of straight hair logistics transportation, improve logistic efficiency.

Step 304, determine the multiple place of departure first nodes to multiple candidate place of departure two-level nodes logistics cost Parameter.

The specific descriptions of the step are referred to above-mentioned steps 202, and details are not described herein again.

Step 305, it logistics list amount and logistics cost parameter is estimated determines mesh according to described using node distribution model Mark place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

The specific descriptions of the step are referred to above-mentioned steps 203, and details are not described herein again.

Optionally, node distribution model is used under conditions of meeting straight hair list amount range, with the smallest each place of departure one Grade node arrives the sum of logistics cost parameter of corresponding place of departure two-level node as target, determine target place of departure two-level node and The corresponding place of departure first nodes of the target place of departure two-level node.

In the embodiment of the present application, by preset node distribution model, each place of departure first nodes are received to correspondence Ground two-level node estimates logistics list amount and each place of departure first nodes to each place of departure for corresponding to place of acceptance two-level node The distance between two-level node, by genetic algorithm, under conditions of meeting straight hair list amount range, with the smallest each place of departure one The sum of the distance of grade node to corresponding place of departure two-level node is target, and it is excellent to distribute at least one for each place of departure first nodes The place of departure two-level node of choosing, place of departure first nodes can be chosen any one from multiple preferred place of departure two-level nodes It is a, or further multiple preferred place of departure two-level nodes can be screened by simulated annealing, it selects optimal Place of departure two-level node.

Optionally, step 305 can also include sub-step 3051, sub-step 3052.

Sub-step 3051 estimates logistics list amount and logistics cost parameter for described, inputs genetic model, determine at least one The corresponding relationship of group target place of departure two-level node and the target place of departure two-level node and place of departure first nodes.

In the embodiment of the present application, node distribution model includes: genetic model.

In the embodiment of the present application, the specific steps of target place of departure two-level node selection are carried out such as based on genetic model Under:

Step A1: construction initial solution.

In this step, for a place of acceptance two-level node, to place of departure first nodes and place of departure two-level node into Row initial code, after coding.Place of departure first nodes become a vector, the digital representation place of departure first nodes in vector Corresponding consolidating the load place of departure two-level node.Usually require generate two initial codes as in genetic algorithm male parent and parent, The fitness of male parent and parent is calculated simultaneously, this fitness is each place of departure two-level node and corresponding each place of departure level-one The sum of the distance between node.

Such as: logistics list amount is estimated to a place of acceptance two-level node by statistics place of departure first nodes, is made and estimates Logistics list scale, while counting place of departure first nodes and place of departure first nodes and delivery is made in place of departure two-level node distance Ground two-level node is apart from table.

Estimate logistics list scale

Place of departure first nodes and place of departure two-level node are apart from table

Equipped with place of departure first nodes collection { A1, A2 ..., An }, place of departure two-level node collection J1, J2 ..., Jm }.

With n=10, (there are ten place of departure first nodes, 3 place of departure two-level nodes) for m=3

Example, an initial solution can be structured as [2,1,3,1,2,2,3,1,1,2], first subscript number of the initial solution Place of departure first nodes A1 is distributed to place of departure two-level node J2 by the expression of word 2, and second index number 1 is indicated place of departure one Grade node A2 distributes to place of departure two-level node J1, and third index number 3 indicates place of departure first nodes A3 distributing to hair Goods ground two-level node J3, and so on.

Therefore, two solution Sol1 and Sol2 can be constructed as parent's body:

Sol1=[2,1,3,1,2,2,3,1,1,2]；

Sol2=[1,2,3,1,1,2,2,2,1,2].

The two solutions represent the corresponding relationship of two kinds of place of departure first nodes and place of departure two-level node, calculate Sol1 and The fitness K1 and K2 of Sol2, i.e., according to place of departure first nodes and place of departure two-level node apart from table, calculate Sol1 and Sol2 respectively the distance between each place of departure two-level node corresponding to corresponding relationship and corresponding each place of departure first nodes it With.

Further, the range of condition for estimating logistics list amount summation for meeting straight hair, such as 9000-12000 can be set.

Step A2: new solution is generated.

In this step, (crossover) and variation (mutation), the strategy of intersection are intersected to male parent and parent The general method exchanged with truncation, the strategy of variation generally with fixed mutation probability or probability of final extinction, to male parent and parent into Row intersects and variation can evolve more solutions, and can evolve more preferred solution, passes through the generation that intersects and make a variation Meet under the conditions of straight hair list amount the sum of the distance between each place of departure two-level node and corresponding each place of departure first nodes out more Small solution, i.e., preferred solution.

For example, intersecting is specially to Sol1=[2,1,3,1,2,2,3,1,1,2]；Sol2=[1,2,3,1,1,2,2,2, 1,2] intersected, selected in Sol1 and Sol2 between the 6th subscript and the 7th subscript as point of cut-off, it will be under the first six digits of Sol1 Rear four subscripts of mark and Sol2 are combined into a new filial generation [2,1,3,1,2,2,2,2,1,2], will be under the first six digits of Sol2 Rear four subscripts of mark and Sol1 are combined into another new filial generation [1,2,3,1,1,2,3,1,1,2].

Variation is makes a variation with random probability to a certain index number in Sol1 and Sol2, in this citing, variation M (m=3) cannot be greater than by obtaining index number afterwards, i.e., no more than the number of place of departure two-level node.

Such as: sol1=[2,1,3,1,2,2,3,1,1,2]

The obtained new filial generation that makes a variation is [1,1,3,1,2,2,3,1,1,2]

Step A3: fine tuning feasible solution.

In this step, for obtained filial generation, by the corresponding place of departure first nodes of filial generation and place of departure two-level node Relationship in, calculate each place of departure two-level node for place of acceptance two-level node and always estimate logistics list amount and each place of departure second level The sum of the distance between node and corresponding each place of departure first nodes, will be unsatisfactory for the sum of the distance of straight hair list amount condition multiplied by One preset penalty coefficient (such as penalty coefficient=1.2), reduces the fitness of the filial generation (with the sum of minimum range for target Under constraint, sum of the distance can make sum of the distance become larger multiplied by a value greater than one, and the distance the big, as the probability preferably solved It is lower), when carrying out cross and variation again in this way, because the filial generation fitness is small, this would not be selected to be unsatisfactory for the filial generation of constraint As parent's body.

Finally, repeating step A2 and A3, iteration preset times select in current sequence the filial generation of fitness highest as male parent And parent, repeated overlapping variation process, until meeting sequence length.

Due to different initial parent's bodies, different optimum solutions may be finally obtained, therefore the production to parent's body here It is raw also to do more wheel calculating.During a new round calculates, after generating initial parent's body, above-mentioned process is repeated, this initial parent is obtained The optimal solution that body finally generates.For example, if desired generating 100 solutions, iteration 100 is needed to take turns.

The last solution that all parent's bodies generate is compared, the maximum solution of fitness is chosen, as final output.By final output Solution, be converted back into the delivery relationship of place of departure first nodes Yu target place of departure two-level node, i.e., to place of departure first nodes point With a target place of departure two-level node, if place of departure first nodes are not previously allocated target place of departure two-level node, still It delivers goods to Distribution Center.

Sub-step 3052, node distribution model further include: annealing model, when there is multiple groups target place of departure two-level node, And the target place of departure two-level node and place of departure first nodes corresponding relationship when, utilize the annealing model, determine One group of optimal target place of departure two-level node and the target place of departure two-level node are corresponding with place of departure first nodes Relationship.

In the embodiment of the present application, it is based on Genetic Algorithm Model, simulated annealing is applied in sub-step 3051, is losing Step A2 and A3 are repeated in propagation algorithm, iteration preset times select in current sequence the filial generation of fitness highest as male parent and mother The step of body.

Simulated annealing provides a kind of acceptance criterion functionIn the function In, T0 is preset initial threshold, and x is the ratio of current iteration number and maximum number of iterations, and α is half-life period, initial threshold Function curve it is as shown in Figure 3A, the number of iterations is bigger, this means that the probability that filial generation is received as parent's body is smaller, iteration time Number is smaller, this means that the probability that filial generation is received as parent's body is bigger, by acceptance criterion function, each iteration can be allowed to produce Raw solution more has an opportunity to jump out local optimum, has advanced optimized the selection of parent's body, so that obtaining most after mostly wheel iteration The optimal solution generated eventually.It should be noted that place of departure first nodes can be selected from multiple preferred place of departure two-level nodes Any one is taken, or genetic algorithm can be helped by Simulated Anneal Algorithm Optimize genetic algorithm to the selection strategy of parent's body Jump out locally optimal solution.The application does not limit this.

Referring to Fig. 4, the step of showing a kind of logistics transportation method that method is determined based on logistics node of the application stream Cheng Tu can specifically include following steps:

Step 401, candidate place of acceptance two-level node is selected from each place of acceptance first nodes according to the first preset rules.

Step 301 can be referred in the step, details are not described herein again.

Step 402, determine that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node.

In this step, each place of departure first nodes arrive the history logistics list of each place of acceptance first nodes within a preset time Amount data are previously stored in the database of logistics system, and server can estimate place of departure by extracting history logistics list amount data First nodes estimate sub- logistics list amount to each place of acceptance first nodes, then the place of acceptance according to belonging to place of acceptance first nodes Two-level node counts multiple place of departure first nodes and estimates logistics list amount to place of acceptance two-level node.

Step 403, candidate place of departure two-level node is selected from each place of departure first nodes according to the second preset rules.

Step 303 can be referred in the step, details are not described herein again.

Step 404, determine the multiple place of departure first nodes to multiple candidate place of departure two-level nodes logistics cost Parameter.

Step 304 can be referred in the step, details are not described herein again.

Step 405, it logistics list amount and logistics cost parameter is estimated determines mesh according to described using node distribution model Mark place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

Step 305 can be referred in the step, details are not described herein again.

Step 406, it is sent to that target place of acceptance two-level node is corresponding to receive when the place of departure first nodes receive When the logistics package of ground first nodes, determine that the stream line of the logistics package is by the place of departure first nodes to correspondence Target place of departure two-level node, then by the target place of departure two-level node to the target place of acceptance two-level node, then by The target place of acceptance two-level node is to the place of acceptance first nodes.

In this step, express delivery reaches any one intermediate transit point in dispatching process and can all be determined by scanning recognition code The current dispatching progress of express delivery, after scanning recognition code, dispatching progress msg can be uploaded to server.

Specifically, a place of departure first nodes are requested by sending poly- goods to server, server is obtained according to the poly- goods It requests the place of departure two-level node of feedback to distribute information, distributes a target place of departure second level section for a place of departure first nodes Point, when pull part person the express delivery for being sent to a consolidating the load region is pulled be sent to a place of departure first nodes when, place of departure first nodes root According to the target place of departure two-level node information of distribution, the dispatching progress msg of express delivery is sent to server, clothes by scanning express delivery The dispatching progress msg of express delivery is fed back to customer by terminal by business device, is target where delivery so that customer understands express delivery Ground two-level node carries out poly- goods.

After target place of departure two-level node receives the express delivery for being sent to the consolidating the load region, by the knowledge for scanning express delivery Other code is simultaneously uploaded to server, can determine that the delivery object of express delivery is corresponding target place of acceptance two-level node, determine mesh After marking the corresponding target place of acceptance two-level node of place of departure two-level node, express delivery is distributed to target place of acceptance two-level node, then By the express delivery received is sent to corresponding place of acceptance first nodes according to place of acceptance by target place of acceptance two-level node, finally It is dispatched into customer's hand by old process.

Step 407, it receives when the place of departure first nodes receive to be sent to other place of acceptance two-level node is corresponding When the logistics package of ground first nodes, then the logistics package is sent in a conventional manner.

In this step, other corresponding place of acceptances of place of acceptance two-level node are sent to when place of departure first nodes receive When the logistics package of first nodes, then it can determine not up to straight hair condition, logistics packet can be sent in a conventional manner at this time It wraps up in, i.e., sends logistics package in such a way that place of departure first nodes are via place of departure relay centre.

In conclusion a kind of logistics node provided by the embodiments of the present application determines method, the application is by each place of acceptance level-one Node is divided into multiple consolidating the load regions by routing rule, and one is selected from each place of acceptance first nodes in each consolidating the load region As the place of acceptance two-level node in the consolidating the load region, for each place of acceptance two-level node, by Density Clustering rule from each delivery Selection is directed to multiple place of departure two-level nodes of the place of acceptance two-level node in ground first nodes；It carries out estimating place of departure later First nodes estimate sub- logistics list amount to each place of acceptance first nodes, are received described according to each place of departure first nodes Ground two-level node estimate logistics list amount and each place of departure first nodes to the place of acceptance two-level node each place of departure The distance between two-level node distributes multiple place of departure two-level nodes in the consolidating the load region for each place of departure first nodes In one, achieved the purpose that increase on the basis of original straight hair route and spelled goods straight hair route, can be by quickly gathering Package to the place of departure two-level node for collecting multiple sender nodes carries out the open-minded of additional straight hair route, and poly- goods is sent out to one Goods the time needed for two-level node it is shorter, solve in the starting point of existing straight hair route that logistics package assemble index is long to ask Topic, increases the quantity of straight hair route, improves logistic efficiency, and estimate logistics list amount and logistics cost ginseng due to combining Number participates in the determination process of logistics node, makes the path for wrapping up poly- goods short as far as possible, so that logistics node is determined to root It is optimized according to logistics cost parameter, there is the beneficial effect for reducing logistics cost.

Referring to Fig. 5, a kind of structure chart of logistics node determining device of the application is shown, can specifically include such as lower die Block:

First determining module 501, for determining that multiple place of departure first nodes estimate logistics to place of acceptance two-level node Single amount

Second determining module 502, for determining the multiple place of departure first nodes to multiple candidate place of departure second level sections The logistics cost parameter of point.

Node distribution module 503 is distributed for estimating logistics list amount and logistics cost parameter according to using node Model determines target place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

Referring to Fig. 6, show a kind of concrete structure diagram of logistics node determining device of the application, can specifically include as Lower module:

First choice module 601, for selecting candidate to receive from each place of acceptance first nodes according to the first preset rules Ground two-level node.

Optionally, first choice module 601 can also include:

Consolidating the load region division submodule, for each place of acceptance first nodes to be divided into multiple shipping areas by routing rule Domain；

First choice submodule, for selecting a history from each place of acceptance first nodes in each consolidating the load region Logistics list amount is greater than the place of acceptance first nodes of the first preset threshold as the candidate place of acceptance second level section for being directed to the consolidating the load region Point.

First determining module 602, for determining that multiple place of departure first nodes estimate logistics to place of acceptance two-level node Dan Liang.

Optionally, the first determining module 602 can also include:

First acquisition submodule, for obtaining the history logistics of the place of departure first nodes to each place of acceptance first nodes Dan Liang.

First estimates submodule, for estimating the place of departure first nodes to each receipts according to the history logistics list amount Goods first nodes estimate sub- logistics list amount.

Optionally, it first estimates submodule and can also include:

Regression model unit, for the place of departure first nodes to be arrived each place of acceptance first nodes within a preset time History logistics list amount and the corresponding factor parameter of history logistics list amount input regression model, obtain the place of departure first nodes and arrive Each place of acceptance first nodes estimate sub- logistics list amount, and the factor parameter includes: inventory information, month, one in tendency Kind is a variety of, the history that the regression model passes through the place of departure first nodes in preset time to each place of acceptance first nodes Logistics list amount and the corresponding factor parameter training of history logistics list amount obtain.

Time series models unit, for the place of departure first nodes to be arrived each place of acceptance level-one section within a preset time The history logistics list amount input time series model of point, obtains the place of departure first nodes to the pre- of each place of acceptance first nodes Estimate sub- logistics list amount, the time series models arrive each place of acceptance level-one by the place of departure first nodes within a preset time The corresponding date training of the history logistics list amount and history logistics list amount of node obtains.

Average value estimates unit, for the place of departure first nodes to be arrived each place of acceptance first nodes within a preset time The average value of history logistics list amount estimate sub- logistics list as the place of departure first nodes to each place of acceptance first nodes Amount.

First estimates unit, for the place of departure first nodes to be arrived each place of acceptance first nodes within a preset time History logistics list amount and the corresponding factor parameter of history logistics list amount input regression model, obtain the place of departure first nodes and arrive The first of each place of acceptance first nodes estimates sub- logistics list amount.

Second estimates unit, for the place of departure first nodes to be arrived each place of acceptance first nodes within a preset time History logistics list amount input time series model, obtain the place of departure first nodes to each place of acceptance first nodes second are pre- Estimate sub- logistics list amount.

Third estimates unit, for the place of departure first nodes to be arrived each place of acceptance first nodes within a preset time The average value of history logistics list amount estimates sub- logistics as the third of the place of departure first nodes to each place of acceptance first nodes Dan Liang.

It is weighted and averaged unit, for estimating sub- logistics list amount for first, second estimating sub- logistics list amount and third estimates son Logistics list amount is weighted and averaged with default weight, obtains place of departure first nodes the estimating to each place of acceptance first nodes Sub- logistics list amount.

Statistic submodule, for counting each hair according to the corresponding place of acceptance first nodes of each candidate's place of acceptance two-level node Goods first nodes estimate logistics list amount to each candidate place of acceptance two-level node.

Second selecting module 603, for selecting candidate send out from each place of departure first nodes according to the second preset rules Goods ground two-level node.

Optionally, the second selecting module 603 can also include:

Site cluster polymerize submodule, for the longitude and latitude according to place of departure first nodes present position, by the hair Goods first nodes be polymerized to multiple site clusters according to clustering algorithm.

Second selection submodule, for sub- logistics list amount will to be estimated in the site cluster greater than the second preset threshold, and, away from The nearest place of departure first nodes in off-network point cluster center, are determined as candidate place of departure two-level node.

Second determining module 604, for determining the multiple place of departure first nodes to multiple candidate place of departure second level sections The logistics cost parameter of point.

Node distribution module 605 is distributed for estimating logistics list amount and logistics cost parameter according to using node Model determines target place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

Optionally, node distribution module 605 can also include:

Genetic model submodule inputs genetic model, really for estimating logistics list amount and logistics cost parameter for described Fixed at least one set target place of departure two-level node and the target place of departure two-level node are corresponding with place of departure first nodes Relationship.

Annealing model submodule, when there is multiple groups target place of departure two-level node and the target place of departure second level section When the corresponding relationship of point and place of departure first nodes, using the annealing model, one group of optimal target place of departure second level is determined The corresponding relationship of node and the target place of departure two-level node and place of departure first nodes.

First path determining module 606, for being sent to target place of acceptance two when the place of departure first nodes receive When the logistics package of the corresponding place of acceptance first nodes of grade node, determine that the stream line of the logistics package is by the delivery Ground first nodes are received to corresponding target place of departure two-level node, then by the target place of departure two-level node to the target Ground two-level node, then by the target place of acceptance two-level node to the place of acceptance first nodes.

Second path determination module 607, for being sent to other place of acceptances two when the place of departure first nodes receive When the logistics package of the corresponding place of acceptance first nodes of grade node, then the logistics package is sent in a conventional manner.

In conclusion a kind of logistics node determining device provided by the embodiments of the present application, the application is by each place of acceptance level-one Node is divided into multiple consolidating the load regions by routing rule, and one is selected from each place of acceptance first nodes in each consolidating the load region As the place of acceptance two-level node in the consolidating the load region, for each place of acceptance two-level node, by Density Clustering rule from each delivery Selection is directed to multiple place of departure two-level nodes of the place of acceptance two-level node in ground first nodes；It carries out estimating place of departure later First nodes estimate sub- logistics list amount to each place of acceptance first nodes, are received described according to each place of departure first nodes Ground two-level node estimate logistics list amount and each place of departure first nodes to the place of acceptance two-level node each place of departure The distance between two-level node distributes multiple place of departure two-level nodes in the consolidating the load region for each place of departure first nodes In one, achieved the purpose that increase on the basis of original straight hair route and spelled goods straight hair route, can be by quickly gathering Package to the place of departure two-level node for collecting multiple sender nodes carries out the open-minded of additional straight hair route, and poly- goods is sent out to one Goods the time needed for two-level node it is shorter, solve in the starting point of existing straight hair route that logistics package assemble index is long to ask Topic, increases the quantity of straight hair route, logistic efficiency can be improved, and estimate logistics list amount and logistics cost due to combining Parameter participates in the determination process of logistics node, makes the path for wrapping up poly- goods short as far as possible, so that logistics node is determined to It is optimized according to logistics cost parameter, there is the beneficial effect for reducing logistics cost.

For device embodiment, since it is basically similar to the method embodiment, related so being described relatively simple Place illustrates referring to the part of embodiment of the method.

Fig. 7 is a kind of structural schematic diagram of server provided by the embodiments of the present application.Referring to Fig. 7, server 700 can be used In the post house address recommended method for implementing to provide in above-described embodiment.The server 700 can be generated because configuration or performance are different Bigger difference may include one or more central processing units (central processing units, CPU) 722 (for example, one or more processors) and memory 732, one or more storage application programs 742 or data 744 storage medium 730 (such as one or more mass memory units).Wherein, memory 732 and storage medium 730 It can be of short duration storage or persistent storage.The program for being stored in storage medium 730 may include one or more moulds Block (diagram does not mark), each module may include to the series of instructions operation in server.Further, central processing Device 722 can be set to communicate with storage medium 730, and the series of instructions behaviour in storage medium 730 is executed on server 700 Make.

Server 700 can also include one or more power supplys 726, one or more wired or wireless networks Interface 750, one or more input/output interfaces 758, one or more keyboards 756, and/or, one or one The above operating system 741, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc.. Wherein, central processing unit 722 can execute the following instruction operated on server 700:

Determine that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node；

Determine the multiple place of departure first nodes to multiple candidate place of departure two-level nodes logistics cost parameter；

Logistics list amount and logistics cost parameter are estimated according to described, using node distribution model, determines target place of departure Two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

Optionally, candidate place of acceptance two-level node is selected from each place of acceptance first nodes according to the first preset rules.

Optionally, each place of acceptance first nodes are divided into multiple consolidating the load regions by routing rule；

From each place of acceptance first nodes in each consolidating the load region, a history logistics list amount is selected to be greater than first default The place of acceptance first nodes of threshold value are as the candidate place of acceptance two-level node for being directed to the consolidating the load region.

Optionally, candidate place of departure two-level node is selected from each place of departure first nodes according to the second preset rules.

Optionally, according to the longitude and latitude of place of departure first nodes present position, the place of departure first nodes are pressed Multiple site clusters are polymerized to according to clustering algorithm；

Sub- logistics list amount will be estimated in the site cluster greater than the second preset threshold, and, nearest apart from site cluster center Place of departure first nodes are determined as candidate place of departure two-level node.

Optionally, obtain the place of departure first nodes to each place of acceptance first nodes history logistics list amount；

According to the history logistics list amount, the place of departure first nodes are estimated to each place of acceptance first nodes and estimate son Logistics list amount；

According to the corresponding place of acceptance first nodes of each candidate's place of acceptance two-level node, each place of departure first nodes are counted to respectively Candidate place of acceptance two-level node estimates logistics list amount.

Optionally, the node distribution model is used under conditions of meeting straight hair list amount range, with the smallest each delivery The sum of the logistics cost parameter of ground first nodes to corresponding place of departure two-level node is target, determines target place of departure two-level node And the corresponding place of departure first nodes of the target place of departure two-level node.

Optionally, the node distribution model includes: genetic model；It is described that logistics list amount and logistics are estimated according to Cost parameter determines target place of departure two-level node and the target place of departure two-level node pair using node distribution model The step of place of departure first nodes answered, comprising:

Logistics list amount and logistics cost parameter are estimated by described, inputs genetic model, determines at least one set target delivery The corresponding relationship of ground two-level node and the target place of departure two-level node and place of departure first nodes.

Optionally, the node distribution model further include: annealing model；It is described to estimate logistics list amount and logistics for described Cost parameter inputs genetic model, determines at least one set target place of departure two-level node and the target place of departure second level section The step of corresponding relationship of point and place of departure first nodes, comprising:

When there is multiple groups target place of departure two-level node and the target place of departure two-level node and place of departure level-one section When the corresponding relationship of point, using the annealing model, one group of optimal target place of departure two-level node and the target are determined The corresponding relationship of place of departure two-level node and place of departure first nodes.

Optionally, the place of departure first nodes are arrived to the history logistics list of each place of acceptance first nodes within a preset time It measures factor parameter corresponding with history logistics list amount and inputs regression model, obtain the place of departure first nodes to each place of acceptance one Grade node estimates sub- logistics list amount, and the factor parameter includes: one of inventory information, month, tendency or a variety of, institute State regression model by the history logistics list amounts of the place of departure first nodes in preset time to each place of acceptance first nodes with The corresponding factor parameter training of history logistics list amount obtains.

Optionally, the place of departure first nodes are arrived to the history logistics list of each place of acceptance first nodes within a preset time Input time series model is measured, the place of departure first nodes is obtained to each place of acceptance first nodes and estimates sub- logistics list amount, The time series models arrive the history object of each place of acceptance first nodes by the place of departure first nodes within a preset time Stream is single to measure date training acquisition corresponding with history logistics list amount.

Optionally, the place of departure first nodes are arrived to the history logistics list of each place of acceptance first nodes within a preset time The average value of amount estimates sub- logistics list amount as the place of departure first nodes to each place of acceptance first nodes.

Optionally, the place of departure first nodes are arrived to the history logistics list of each place of acceptance first nodes within a preset time It measures factor parameter corresponding with history logistics list amount and inputs regression model, obtain the place of departure first nodes to each place of acceptance one The first of grade node estimates sub- logistics list amount；

The place of departure first nodes are arrived to the history logistics list amount input of each place of acceptance first nodes within a preset time Time series models obtain the place of departure first nodes to the second of each place of acceptance first nodes and estimate sub- logistics list amount；

The place of departure first nodes are arrived into the flat of the history logistics list amount of each place of acceptance first nodes within a preset time Mean value estimates sub- logistics list amount as the third of the place of departure first nodes to each place of acceptance first nodes；

Sub- logistics list amount is estimated by first, second estimates sub- logistics list amount and third estimates sub- logistics list amount to preset weight It is weighted and averaged, obtains the place of departure first nodes to each place of acceptance first nodes and estimate sub- logistics list amount.

Optionally, the corresponding place of acceptance of target place of acceptance two-level node is sent to when the place of departure first nodes receive When the logistics package of first nodes, the stream line for determining logistics package is by the place of departure first nodes to corresponding Target place of departure two-level node, then by the target place of departure two-level node to the target place of acceptance two-level node, then by institute Target place of acceptance two-level node is stated to the place of acceptance first nodes；

The corresponding place of acceptance level-one section of other place of acceptance two-level nodes is sent to when the place of departure first nodes receive When the logistics package of point, then the logistics package is sent in a conventional manner.

The embodiment of the present application provides a kind of device, is stored thereon with one or more machine readable medias of instruction, when by When one or more of processors execute described instruction, so that described device executes a kind of logistics node and determines method.

The embodiment of the present application also provides one or more machine readable medias, instruction is stored thereon with, when by one or more When a processor executes described instruction, executes a kind of logistics node and determine method.

All the embodiments in this specification are described in a progressive manner, the highlights of each of the examples are with The difference of other embodiments, the same or similar parts between the embodiments can be referred to each other.

It should be understood by those skilled in the art that, the embodiments of the present application may be provided as method, apparatus or calculating Machine program product.Therefore, the embodiment of the present application can be used complete hardware embodiment, complete software embodiment or combine software and The form of the embodiment of hardware aspect.Moreover, the embodiment of the present application can be used one or more wherein include computer can With in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of program code The form of the computer program product of implementation.

The embodiment of the present application is referring to according to the method for the embodiment of the present application, terminal device (system) and computer program The flowchart and/or the block diagram of product describes.It should be understood that flowchart and/or the block diagram can be realized by computer program instructions In each flow and/or block and flowchart and/or the block diagram in process and/or box combination.It can provide these Computer program instructions are set to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing terminals Standby processor is to generate a machine, so that being held by the processor of computer or other programmable data processing terminal devices Capable instruction generates for realizing in one or more flows of the flowchart and/or one or more blocks of the block diagram The device of specified function.

These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing terminal devices In computer-readable memory operate in a specific manner, so that instruction stored in the computer readable memory generates packet The manufacture of command device is included, which realizes in one side of one or more flows of the flowchart and/or block diagram The function of being specified in frame or multiple boxes.

These computer program instructions can also be loaded into computer or other programmable data processing terminal devices, so that Series of operation steps are executed on computer or other programmable terminal equipments to generate computer implemented processing, thus The instruction executed on computer or other programmable terminal equipments is provided for realizing in one or more flows of the flowchart And/or in one or more blocks of the block diagram specify function the step of.

Although preferred embodiments of the embodiments of the present application have been described, once a person skilled in the art knows bases This creative concept, then additional changes and modifications can be made to these embodiments.So the following claims are intended to be interpreted as Including preferred embodiment and all change and modification within the scope of the embodiments of the present application.

Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning Covering non-exclusive inclusion, so that process, method, article or terminal device including a series of elements not only wrap Those elements are included, but also including other elements that are not explicitly listed, or further includes for this process, method, article Or the element that terminal device is intrinsic.In the absence of more restrictions, being wanted by what sentence "including a ..." limited Element, it is not excluded that there is also other identical elements in process, method, article or the terminal device for including the element.

Method and device is determined to a kind of logistics node provided herein above, is described in detail, herein Applying specific case, the principle and implementation of this application are described, and the explanation of above example is only intended to help Understand the present processes and its core concept；At the same time, for those skilled in the art, according to the thought of the application, There will be changes in the specific implementation manner and application range, in conclusion the content of the present specification should not be construed as to this The limitation of application.

Claims (30)

1. a kind of logistics node determines method, which is characterized in that including

Determine that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node；

Determine the multiple place of departure first nodes to multiple candidate place of departure two-level nodes logistics cost parameter；

Logistics list amount and logistics cost parameter are estimated according to described, using node distribution model, determines target place of departure second level Node and the corresponding place of departure first nodes of the target place of departure two-level node.

2. the method according to claim 1, wherein further include:

Candidate place of acceptance two-level node is selected from each place of acceptance first nodes according to the first preset rules.

3. according to the method described in claim 2, it is characterized in that, it is described according to the first preset rules from each place of acceptance level-one section The step of candidate place of acceptance two-level node is selected in point, comprising:

Each place of acceptance first nodes are divided into multiple consolidating the load regions by routing rule；

From each place of acceptance first nodes in each consolidating the load region, a history logistics list amount is selected to be greater than the first preset threshold Place of acceptance first nodes as be directed to the consolidating the load region candidate place of acceptance two-level node.

4. the method according to claim 1, wherein further include:

Candidate place of departure two-level node is selected from each place of departure first nodes according to the second preset rules.

5. according to the method described in claim 4, it is characterized in that, it is described according to the second preset rules from each place of departure level-one The step of candidate place of departure two-level node is selected in node, comprising:

According to the longitude and latitude of place of departure first nodes present position, the place of departure first nodes are gathered according to clustering algorithm It is combined into multiple site clusters；

Sub- logistics list amount will be estimated in the site cluster greater than the second preset threshold, and, the delivery nearest apart from site cluster center Ground first nodes are determined as candidate place of departure two-level node.

6. the method according to claim 1, wherein the multiple place of departure first nodes of the determination are to place of acceptance two The step of estimating logistics list amount of node of grade, comprising:

Obtain the place of departure first nodes to each place of acceptance first nodes history logistics list amount；

According to the history logistics list amount, the place of departure first nodes are estimated to each place of acceptance first nodes and estimate sub- logistics Dan Liang；

According to the corresponding place of acceptance first nodes of each candidate's place of acceptance two-level node, each place of departure first nodes are counted to each candidate Place of acceptance two-level node estimates logistics list amount.

7. the method according to claim 1, wherein the node distribution model is for meeting straight hair list amount model Under conditions of enclosing, with the sum of the smallest each place of departure first nodes to the logistics cost parameter of corresponding place of departure two-level node for mesh Mark, determines target place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

8. the method according to the description of claim 7 is characterized in that the node distribution model includes: genetic model；Described Estimate logistics list amount and logistics cost parameter according to described, using node distribution model, determine target place of departure two-level node with And the step of target place of departure two-level node corresponding place of departure first nodes, comprising:

Logistics list amount and logistics cost parameter are estimated by described, inputs genetic model, determines at least one set target place of departure two The corresponding relationship of grade node and the target place of departure two-level node and place of departure first nodes.

9. according to the method described in claim 8, it is characterized in that, the node distribution model further include: annealing model；It is described Logistics list amount and logistics cost parameter are estimated by described, inputs genetic model, determines at least one set target place of departure second level section The step of corresponding relationship of point and the target place of departure two-level node and place of departure first nodes, comprising:

When there are multiple groups target place of departure two-level node and the target place of departure two-level node and place of departure first nodes When corresponding relationship, using the annealing model, optimal one group of target place of departure two-level node and target delivery are determined The corresponding relationship of ground two-level node and place of departure first nodes.

10. according to the method described in claim 6, estimating the hair it is characterized in that, described according to the history logistics list amount Goods first nodes to each place of acceptance first nodes the step of estimating sub- logistics list amount, comprising:

The place of departure first nodes are arrived to the history logistics list amount and history object of each place of acceptance first nodes within a preset time Stream is single to measure corresponding factor parameter input regression model, obtains the place of departure first nodes to the pre- of each place of acceptance first nodes Estimate sub- logistics list amount, the factor parameter includes: one of inventory information, month, tendency or a variety of, the regression model Pass through the history logistics list amount and history logistics list of place of departure first nodes described in preset time to each place of acceptance first nodes Corresponding factor parameter training is measured to obtain.

11. according to the method described in claim 6, estimating the hair it is characterized in that, described according to the history logistics list amount Goods first nodes to each place of acceptance first nodes the step of estimating sub- logistics list amount, comprising:

The place of departure first nodes are arrived to the history logistics list amount input time of each place of acceptance first nodes within a preset time Series model obtains estimate sub- logistics list amount of the place of departure first nodes to each place of acceptance first nodes, the time sequence Column model arrives the history logistics list amount of each place of acceptance first nodes by the place of departure first nodes within a preset time and goes through The corresponding date training of history logistics list amount obtains.

12. according to the method described in claim 6, estimating the hair it is characterized in that, described according to the history logistics list amount Goods first nodes to each place of acceptance first nodes the step of estimating sub- logistics list amount, comprising:

The place of departure first nodes are arrived to the average value of the history logistics list amount of each place of acceptance first nodes within a preset time Sub- logistics list amount is estimated as the place of departure first nodes to each place of acceptance first nodes.

13. according to the method described in claim 6, estimating the hair it is characterized in that, described according to the history logistics list amount Goods first nodes to each place of acceptance first nodes the step of estimating sub- logistics list amount, comprising:

The place of departure first nodes are arrived to the history logistics list amount and history object of each place of acceptance first nodes within a preset time Stream is single to be measured corresponding factor parameter and inputs regression model, obtains the place of departure first nodes to the of each place of acceptance first nodes One estimates sub- logistics list amount；

The place of departure first nodes are arrived to the history logistics list amount input time of each place of acceptance first nodes within a preset time Series model obtains the place of departure first nodes to the second of each place of acceptance first nodes and estimates sub- logistics list amount；

The place of departure first nodes are arrived to the average value of the history logistics list amount of each place of acceptance first nodes within a preset time Third as the place of departure first nodes to each place of acceptance first nodes estimates sub- logistics list amount；

Sub- logistics list amount is estimated by first, second estimates sub- logistics list amount and third is estimated sub- logistics list amount and carried out with default weight Weighted average, obtains the place of departure first nodes to each place of acceptance first nodes and estimates sub- logistics list amount.

14. the method according to claim 1, wherein further include:

The corresponding place of acceptance first nodes of target place of acceptance two-level node are sent to when the place of departure first nodes receive When logistics is wrapped up, determine that the stream line of the logistics package is by the place of departure first nodes to corresponding target place of departure Two-level node, then by the target place of departure two-level node to the target place of acceptance two-level node, then received by the target Ground two-level node is to the place of acceptance first nodes；

The corresponding place of acceptance first nodes of other place of acceptance two-level nodes are sent to when the place of departure first nodes receive When logistics is wrapped up, then the logistics package is sent in a conventional manner.

15. a kind of logistics node determining device characterized by comprising

First determining module, for determining that multiple place of departure first nodes estimate logistics list amount to place of acceptance two-level node；

Second determining module, for determining the logistics of the multiple place of departure first nodes to multiple candidate place of departure two-level nodes Cost parameter；

Node distribution module, for estimating logistics list amount and logistics cost parameter according to, using node distribution model, really Set the goal place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

16. device according to claim 15, which is characterized in that further include:

First choice module, for selecting candidate place of acceptance second level section from each place of acceptance first nodes according to the first preset rules Point.

17. device according to claim 16, which is characterized in that the first choice module, comprising:

Consolidating the load region division submodule, for each place of acceptance first nodes to be divided into multiple consolidating the load regions by routing rule；

First choice submodule, for selecting a history logistics from each place of acceptance first nodes in each consolidating the load region Single amount is greater than the place of acceptance first nodes of the first preset threshold as the candidate place of acceptance two-level node for being directed to the consolidating the load region.

18. device according to claim 15, which is characterized in that further include:

Second selecting module, for selecting candidate place of departure second level from each place of departure first nodes according to the second preset rules Node.

19. device according to claim 18, which is characterized in that second selecting module, comprising:

Site cluster polymerize submodule, for the longitude and latitude according to place of departure first nodes present position, by the place of departure First nodes are polymerized to multiple site clusters according to clustering algorithm；

Second selection submodule, for sub- logistics list amount will to be estimated in the site cluster greater than the second preset threshold, and, distance webs The nearest place of departure first nodes in point cluster center, are determined as candidate place of departure two-level node.

20. device according to claim 15, which is characterized in that first determining module, comprising:

First acquisition submodule, the history logistics list for obtaining the place of departure first nodes to each place of acceptance first nodes Amount；

First estimates submodule, for estimating the place of departure first nodes to each place of acceptance according to the history logistics list amount First nodes estimate sub- logistics list amount；

Statistic submodule, for counting each place of departure according to the corresponding place of acceptance first nodes of each candidate's place of acceptance two-level node First nodes estimate logistics list amount to each candidate place of acceptance two-level node.

21. device according to claim 15, which is characterized in that the node distribution model is for meeting straight hair list amount Under conditions of range, the sum of the logistics cost parameter with the smallest each place of departure first nodes to corresponding place of departure two-level node is Target determines target place of departure two-level node and the corresponding place of departure first nodes of the target place of departure two-level node.

22. device according to claim 21, which is characterized in that the node distribution model includes: genetic model, described Node distribution module, comprising:

Genetic model submodule inputs genetic model, determines extremely for estimating logistics list amount and logistics cost parameter for described Few one group of target place of departure two-level node and target place of departure two-level node pass corresponding with place of departure first nodes System.

23. device according to claim 22, which is characterized in that the node distribution model further include: annealing model, institute State distribution module, comprising:

Annealing model submodule, when occur multiple groups target place of departure two-level node and the target place of departure two-level node with When the corresponding relationship of place of departure first nodes, using the annealing model, one group of optimal target place of departure two-level node is determined, And the corresponding relationship of the target place of departure two-level node and place of departure first nodes.

24. device according to claim 22, which is characterized in that described first estimates submodule, comprising:

Regression model unit, for the place of departure first nodes to be arrived to the history of each place of acceptance first nodes within a preset time Logistics list amount and the corresponding factor parameter of history logistics list amount input regression model, obtain the place of departure first nodes to each receipts Goods first nodes estimate sub- logistics list amount, the factor parameter include: one of inventory information, month, tendency or History logistics a variety of, that the regression model passes through the place of departure first nodes in preset time to each place of acceptance first nodes It is single to measure factor parameter training acquisition corresponding with history logistics list amount.

25. device according to claim 22, which is characterized in that described first estimates submodule, comprising:

Time series models unit, for the place of departure first nodes to be arrived each place of acceptance first nodes within a preset time History logistics list amount input time series model, obtains the place of departure first nodes to each place of acceptance first nodes and estimates son Logistics list amount, the time series models arrive each place of acceptance first nodes by the place of departure first nodes within a preset time History logistics list amount and the training of history logistics list amount corresponding date obtain.

26. device according to claim 22, which is characterized in that described first estimates submodule, comprising:

Average value estimates unit, for the place of departure first nodes to be arrived going through for each place of acceptance first nodes within a preset time The average value of history logistics list amount estimates sub- logistics list amount as the place of departure first nodes to each place of acceptance first nodes.

27. device according to claim 22, which is characterized in that described first estimates submodule, comprising:

First estimates unit, for the place of departure first nodes to be arrived to the history of each place of acceptance first nodes within a preset time Logistics list amount and the corresponding factor parameter of history logistics list amount input regression model, obtain the place of departure first nodes to each receipts Goods first nodes first estimate sub- logistics list amount；

Second estimates unit, for the place of departure first nodes to be arrived to the history of each place of acceptance first nodes within a preset time Logistics list amount input time series model, obtain the place of departure first nodes to each place of acceptance first nodes second estimate son Logistics list amount；

Third estimates unit, for the place of departure first nodes to be arrived to the history of each place of acceptance first nodes within a preset time The average value of logistics list amount estimates sub- logistics list amount as the third of the place of departure first nodes to each place of acceptance first nodes；

It is weighted and averaged unit, for estimating sub- logistics list amount for first, second estimating sub- logistics list amount and third estimates sub- logistics Single amount is weighted and averaged with default weight, is obtained the place of departure first nodes to each place of acceptance first nodes and is estimated sub- object The single amount of stream.

28. device according to claim 15, which is characterized in that further include:

First path determining module, for being sent to target place of acceptance two-level node pair when the place of departure first nodes receive When the logistics package for the place of acceptance first nodes answered, determine that the stream line of the logistics package is by the place of departure level-one section Point arrives corresponding target place of departure two-level node, then by the target place of departure two-level node to the target place of acceptance second level section Point, then by the target place of acceptance two-level node to the place of acceptance first nodes；

Second path determination module, for being sent to other place of acceptance two-level nodes pair when the place of departure first nodes receive When the logistics package for the place of acceptance first nodes answered, then the logistics package is sent in a conventional manner.

29. a kind of device characterized by comprising

One or more processors；With

One or more machine readable medias of instruction are stored thereon with, when being executed by one or more of processors, are made Obtain the method that described device executes such as claim 1-14 one or more.

30. one or more machine readable medias, are stored thereon with instruction, when executed by one or more processors, so that Device executes the method such as claim 1-14 one or more.