CN109508862A - A kind of allocator and device - Google Patents
A kind of allocator and device Download PDFInfo
- Publication number
- CN109508862A CN109508862A CN201811180185.3A CN201811180185A CN109508862A CN 109508862 A CN109508862 A CN 109508862A CN 201811180185 A CN201811180185 A CN 201811180185A CN 109508862 A CN109508862 A CN 109508862A
- Authority
- CN
- China
- Prior art keywords
- dispatching person
- order
- matching
- information
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0832—Special goods or special handling procedures, e.g. handling of hazardous or fragile goods
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application discloses a kind of method and apparatus of intelligent distribution worksheet processing, comprising: obtains dispatching person's information and order information；According to machine learning method, the order probability of dispatching person is calculated using dispatching person's information and order information；Global optimization's matching is carried out according to the order probability of the dispatching person, to determine the matching relationship of order and dispatching person；Order is distributed to the matched dispatching person of global optimum according to the matching relationship.Using technical solution disclosed in the present application, optimum allocation effect is considered due to more precisely calculating order probability using machine learning method, and from global angle, so that maximal efficiency utilizes resource, optimizes Order splitting.
Description
Technical field
This application involves logistics technology, in particular to a kind of allocator and device.
Background technique
With the development of social informatization technology, people are influenced increasing by internet, and many economic activities are all logical
Cross online trading.Especially common daily life is also tended to through network trading, for example is wrapped up express delivery, made a reservation.To make a reservation
For dispatching, since client's flow point dissipates, the features such as real-time requirement degree is high, and required time section is concentrated, how rationally to utilize resource,
Optimizing Order splitting, improving dispatching efficiency to meet customer need is current problem to be solved.
Summary of the invention
The embodiment of the present application provides a kind of method of intelligent distribution worksheet processing, and global optimization's matching, optimization may be implemented
The distribution and utilization of resource.Specifically, the embodiment of the present application are as follows:
A kind of method of intelligent distribution worksheet processing, this method comprises:
Obtain dispatching person's information and order information；
According to machine learning method, the order probability of dispatching person is calculated using dispatching person's information and order information；
Global optimization's matching is carried out according to the order probability of the dispatching person, to determine that the matching of order and dispatching person is closed
System；
Order is distributed to the matched dispatching person of global optimum according to the matching relationship.
Further,
Dispatching person's information includes dispatching person's location information, and the order information includes shipping and receiving location information.
Further, described according to machine learning method, dispatching person is calculated using dispatching person's information and order information
The method of order probability include:
Setting feature weight vector sum bias in advance；
The feature weight vector is updated according to the loss function of setting and feature weight vector iterative calculation；
The order probability of dispatching person is calculated using the feature weight vector sum bias in Growth Function.
Further, global optimization's matching is carried out according to the order probability of the dispatching person, to determine order and dispatching
Member matching relationship method include:
The order probability of the dispatching person is weighted according to matching weighted value set in advance, obtains dispatching person
Match assessed value；
Matching total value is calculated according to the matching assessed value of the dispatching person and allocation strategy score value, most by matching total value
Corresponding allocation strategy determines order and dispatching person according to the allocation strategy of optimization as the allocation strategy optimized when high
Matching relationship.
The embodiment of the present application also proposes that a kind of device of intelligent distribution worksheet processing, the device include:
Information acquisition unit, for obtaining dispatching person's information and order information；
Order probability prediction unit, for utilizing dispatching person's information and order information meter according to machine learning method
Calculate the order probability of dispatching person；
Matching unit carries out global optimization's matching according to the order probability of the dispatching person, to determine order and dispatching
The matching relationship of member；
Order is distributed to the matched dispatching person of global optimum according to the matching relationship by order dispatch unit.
Further, the order probability prediction unit includes:
Connecing for dispatching person is calculated using feature weight vector sum bias in Growth Function in probability calculation subelement
Single probability；
Feature weight vector iteration subelement iterates to calculate more according to the loss function of setting and the feature weight vector
The new feature weight vector.
Further, the matching unit includes:
Assessment unit is matched, for adding according to matching weighted value set in advance to the order probability of the dispatching person
Power calculates, and obtains dispatching person and matches assessed value；
Global optimum's matching primitives unit calculates matching according to the matching assessed value of the dispatching person and allocation strategy score value
Total value, corresponding allocation strategy is as the allocation strategy of optimization when will match total value highest, according to the distribution of optimization
Strategy determines the matching relationship of order and dispatching person.
It is general due to more precisely calculating order using machine learning method using technical solution disclosed in the present application
Rate, and optimum allocation effect is considered from global angle, so that maximal efficiency utilizes resource, optimize Order splitting, realizes intelligence group
It is single.
Detailed description of the invention
Fig. 1 is the method flow diagram of the intelligent distribution worksheet processing of the embodiment of the present invention one.
Fig. 2 is the corresponding apparatus structure schematic diagram of the embodiment of the present invention one.
Fig. 3 is the method flow diagram that dispatching person's order probability is calculated in the embodiment of the present invention two.
Fig. 4 is the schematic diagram of internal structure of order probability prediction unit 202 in the embodiment of the present invention two.
Fig. 5 is to talk about the method flow diagram that matching process determines matching relationship according to global optimum in the embodiment of the present invention two.
Fig. 6 is 203 schematic diagram of internal structure of matching unit in the embodiment of the present invention two.
Specific embodiment
It is right hereinafter, referring to the drawings and the embodiments, for the objects, technical solutions and advantages of the application are more clearly understood
The application is described in further detail.
Fig. 1 is the method flow diagram of the intelligent distribution worksheet processing of the embodiment of the present invention one.As shown in Figure 1, this method comprises:
Step 101: obtaining dispatching person's information and order information.
Step 102: according to machine learning method, the order of dispatching person is calculated using dispatching person's information and order information
Probability.
Step 103: global optimization's matching being carried out according to the order probability of the dispatching person, to determine order and dispatching person
Matching relationship.
Step 104: order is distributed to by the matched dispatching person of global optimum according to the matching relationship.
Fig. 2 is the corresponding apparatus structure schematic diagram of the embodiment of the present invention one.As shown in Fig. 2, the device includes: acquisition of information
Unit 201, order probability prediction unit 202, matching unit 203, order dispatch unit 204.Wherein, information acquisition unit 201
For obtaining dispatching person's information and order information；Order probability prediction unit 202 is used for according to machine learning method, using described
Dispatching person's information and order information calculate the order probability of dispatching person；Matching unit 203 is according to the order probability of the dispatching person
Global optimization's matching is carried out, to determine the matching relationship of order and dispatching person；Order dispatch unit 204 is closed according to the matching
Order is distributed to the matched dispatching person of global optimum by system.
That is, can use the order probability that machine learning method calculates dispatching person in the embodiment of the present invention one,
Dispatching person and order are matched using this order probability, by the matched dispatching person of global optimum as needing to issue order
Dispatching person.It can be client, dispatching person and several sides of businessman since the method for the present embodiment worksheet processing considers global optimum's effect
Face is comprehensively considered, and can be utilized resource with maximal efficiency, be optimized Order splitting.Certainly, due to optimizing Order splitting, thus
Dispatching efficiency can be improved on the whole.
In practical application, dispatching person's information may include its location information, and order information may include shipping and receiving position letter
Breath.Location information can be indicated with longitude and latitude, can be obtained with existing GPS technology.
Certainly, dispatching person's information can also include other information, such as: it is whether in running order, dispatching person's order is inclined
Good, dispatching person's grade, dispatching person's label, dispatching person's order permission, dispatching person's backlog situation, dispatching person's current time push away
Order volume, the shielded situation of dispatching person, the nearest history order situation of dispatching person etc. completed on the day of the order volume recommended, dispatching person
Deng.
Order information can also include other information, such as: the timeliness of order, dispatching person's backlog will be distributed
Timeliness, businessman's ID number of order, order cargo type.
It as which information dispatching person's information and order information include actually is implemented by the application present invention in practical application
What the user of example scheme voluntarily determined, it is not limited by the embodiment of the present invention.
Since dispatching person (M) and the quantity of order (N number of) are very huge, can filtration fraction irrelevant information in advance, reduction
Calculation amount.For example, matching relationship can be preliminarily formed first by M dispatching person and N number of order cross match.Then according to acquisition
Dispatching person's information and order information carry out information sifting, by following situation it is relevant matching filter out: be not currently in work
The dispatching person of state, the dispatching person for not receiving the order cargo type, the unmatched order of order preference, path deviation are big to be ordered
Single, dispatching person being saturated by the dispatching person, the dispatching person without order permission, current order amount of businessman's shielding etc..These are believed
Breath filtering the purpose is to reduce calculation amounts.Certainly, it if not considering the problems of calculation amount, can not also filter in advance.
The embodiment of the present invention two is how to realize that above-mentioned steps 102 provide the scheme of specific machine learning.This field
Technical staff knows that machine learning is a multi-field cross discipline, is related to probability theory, statistics, Approximation Theory, convextiry analysis, calculation
The multiple subjects such as method complexity computation.The learning behavior that the mankind were simulated or realized to computer how is specialized in, to obtain newly
Knowledge or skills reorganize the existing structure of knowledge and are allowed to constantly improve the performance of itself.Fig. 3 is the tool for realizing step 102
Body method flow chart.As shown in figure 3, the method that machine learning calculates dispatching person's order probability in this method includes:
Step 301: setting feature weight vector sum bias in advance.
Step 302: the feature weight is updated according to the loss function of setting and feature weight vector iterative calculation
Vector.
Step 303: the order of dispatching person is calculated using the feature weight vector sum bias in Growth Function
Probability.
Feature weight vector described here, bias, loss function, Growth Function are the skills being related in machine learning
Art feature.Fig. 4 is the corresponding schematic device of this method, the i.e. schematic diagram of internal structure of order probability prediction unit 202.Such as figure
Shown in 4, which includes: probability calculation subelement 401 and feature weight vector iteration subelement 402.Wherein, probability calculation
Unit 401 is used to be calculated the order probability of dispatching person using feature weight vector sum bias in Growth Function.Feature
Weight vectors iteration subelement 402 updates the feature according to the loss function of setting and feature weight vector iterative calculation
Weight vectors.
Here feature weight vector is weight set by the various feature objects that need to learn in machine learning, biasing
Value is a pre-set numerical value, and loss function is needed when machine learning forms stable model, and Growth Function is to use
To calculate the model of order probability.
Assuming that can use in machine learning model training, there are many feature object, such as above-mentioned various dispatching person's information
And order information.The model of machine learning also has very much, such as Logistic, XGBoost etc..Below with Logistic model
For illustrate how to be trained.
When choosing the feature object of training, usually it is contemplated that the variable's attribute of object.Such as: time variable is one
Composite variable may include date and hour, such as in Augusts, 2017 and 18 points 10 minutes.Address variable is composite variable, be may include
Address name and longitude and latitude.Address name is discrete random variable, and longitude and latitude is continuous variable, and the time is periodically to become
Amount.It usually can choose a main feature, how to select feature can be by measuring this feature and by variable to be predicted
Maximally related situation determines.The index for measuring correlation between variable can be embodied by Pearson came system and mutual information.Its
In, correlation of the Pearson came system commonly used to measure two continuous random variables is calculating the result is that linearly related
's.If existing continuity, and have discrete type, preiodic type or compound variable, then it can choose mutual information to measure variable
Between correlation.
In the Logistic model of the present embodiment, it is assumed that trained feature object is needed to be dispatching person's location information, connect
Single time and these three variables of departure date.Dispatching person's order probability is set as y, and dispatching person position is x1, time of received orders x2,
Departure date is x3, feature weight vector is WT, bias b.
So, available according to Logistic model:
Y=H (X)=Sigmoid (WTX+b) (formula 1)
Wherein, for sample (x1, y1), (x2, y2) ... (xn, yn), wherein (0,1) y ∈, indicates x1As spy
Order probability y is obtained when sign object training1, by x2Order probability y is obtained when as feature object training2, by x3As feature pair
As obtaining order probability y when training3。H(X)(y1, x1) indicate x1And y1Between correlation, H (X) (y2, x2) indicate x2And y2It
Between correlation, H (X) (y3, x3) indicate x3And y3Between correlation.
Assuming that obtaining result as shown in Table 1 after calculating:
| H(X)(y1, x1) | H(X)(y2, x2) | H(X)(y3, x3) |
| 0.91 | 0.43 | 0.82 |
Table one
From table one it is found that, obtained order probability and x1Correlation it is most strong, reached 0.91.Therefore, it is actually answering
Dispatching person position x may be selected in1Best features as training pattern.That is, can be by problem reduction are as follows: calculate
It meets dispatching person position under the conditions of order in other words for the undertaking probability of the order under this feature of dispatching person position
Probability distribution.
Loss function is introduced in a model:
It is constantly iterated using formula 1 and formula 2:
Wherein, WoldWeight vectors before indicating iteration, WnewWeight vectors after indicating iteration.By continuous iteration,
The available weight vectors to tend towards stability.In this way, the Logistic model by formula 1 calculates, so that it may obtain reliable
Dispatching person's order probability.
The embodiment of the present invention two is how to realize that above-mentioned steps 103 provide specific match orders and dispatching person's relationship
Method.Fig. 5 is the specific method flow chart for realizing step 103.As shown in figure 3, talking about match party according to global optimum in this method
The method that method determines matching relationship includes:
Step 501: it is weighted according to order probability of the dispatching person's weighted value set in advance to the dispatching person,
It obtains dispatching person and matches assessed value.
In practical application, when calculating matching assessed value, it is also contemplated that dispatching person's information.As previously mentioned, dispatching person's information
It may include this kind of information for embodying dispatching person's business achievement of dispatching person's grade, the business of dispatching person's history can be utilized here
Achievement calculates matching score value.Specifically, matching weighted value can be respectively set for order probability and dispatching person's information, then carry out
Weighted average, to determine the matching score value of dispatching person.Such as: the order probability of certain dispatching person A is 0.8, and corresponding weighted value is
18, grade is " king ", remembers 5 points, and corresponding weighted value is 1, then the matching score value that the dispatching person A is obtained should be 0.8*18
+ 5*1=19.5 points.Here the business achievement of dispatching person is considered, in practical application, it is also contemplated that other information or only examining
Consider order probability.
Step 502: matching total value being calculated according to the matching assessed value of the dispatching person and allocation strategy score value, will be matched
When total value highest corresponding allocation strategy as optimize allocation strategy, according to the allocation strategy of optimization determine order and
The matching relationship of dispatching person.
The matching score value that matching score value, that is, step 501 described here is calculated, allocation strategy score value are a two-values
The value of change is assigned as 1, is not assigned as 0.Assuming that: x1, x2... xMI ∈ { 1...M } indicates M dispatching person；y1,
y2... yNJ ∈ { 1...N } indicates N number of order；αijIndicate the matching score value between dispatching person i and order j；βijIt indicates
Allocation strategy score value, βij=1 indicates order j distributing to dispatching person i, βij=0 indicates order j not to be distributed to dispatching person i.
So, objective function f can be indicated are as follows:
F=max ∑i∑jαijβij(formula 4)
(constraint:K is parameter)
That is, passing through ∑i∑jαijβijFormula can calculate all Order splittings to the matching of different dispatching persons
Total value, and max ∑i∑jαijβijIt is then highest score in several matching total values, has corresponded to a kind of optimization in other words
Allocation strategy.The allocation strategy of this optimization has corresponded to the matching relationship that some order j is distributed to some dispatching person i,
It is a kind of matching relationship of optimization.
Fig. 6 is the corresponding device figure of determining matching relationship, i.e. 203 schematic diagram of internal structure of matching unit.As shown in fig. 6,
The device includes matching assessment unit 601 and global optimum's matching primitives unit 602.
Wherein, assessment unit 601 is matched, for general to the order of the dispatching person according to matching weighted value set in advance
Rate is weighted, and obtains dispatching person and matches assessed value.Global optimum's matching primitives unit 602, according to the dispatching person's
It matches assessed value and allocation strategy score value and calculates matching total value, corresponding allocation strategy will be matched when total value highest as most
The allocation strategy of optimization determines order and the matching relationship of dispatching person according to the allocation strategy of optimization.Determine optimal matching
After relationship, so that it may order are distributed to the matched dispatching person of global optimum, realize dispatching task.
The embodiment of the present invention is not single to consider whether from dispatching person's order probability or other information by some order
Some dispatching person is distributed to, but considers the reasonability of distribution from global angle, so that whole system is whithin a period of time
Operation be most rationally, optimize.
Using various embodiments of the present invention scheme, since the method using machine learning can more precisely calculate dispatching
Member accepts the probability of order, and is matched dispatching person and order by global optimization's matching process, thus on the whole
Resource is utilized in maximum efficiency.
The foregoing is merely the preferred embodiments of the application, not to limit the application, all essences in the application
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the application protection.
Claims (7)
1. a kind of method of intelligent distribution worksheet processing, which is characterized in that this method comprises:
Obtain dispatching person's information and order information；
According to machine learning method, the order probability of dispatching person is calculated using dispatching person's information and order information；
Global optimization's matching is carried out according to the order probability of the dispatching person, to determine the matching relationship of order and dispatching person；
Order is distributed to the matched dispatching person of global optimum according to the matching relationship.
2. the method according to claim 1, wherein
Dispatching person's information includes dispatching person's location information, and the order information includes shipping and receiving location information.
3. utilizing the dispatching person the method according to claim 1, wherein described according to machine learning method
The method that information and order information calculate the order probability of dispatching person includes:
Setting feature weight vector sum bias in advance；
The feature weight vector is updated according to the loss function of setting and feature weight vector iterative calculation；
The order probability of dispatching person is calculated using the feature weight vector sum bias in Growth Function.
4. the method according to claim 1, wherein carrying out global optimum according to the order probability of the dispatching person
Change matching, to determine that the matching relationship method of order and dispatching person includes:
The order probability of the dispatching person is weighted according to matching weighted value set in advance, obtains dispatching person's matching
Assessed value；
Matching total value is calculated according to the matching assessed value of the dispatching person and allocation strategy score value, when by matching total value highest
Corresponding allocation strategy determines order and the matching of dispatching person according to the allocation strategy of optimization as the allocation strategy optimized
Relationship.
5. a kind of device of intelligent distribution worksheet processing, which is characterized in that the device includes:
Information acquisition unit, for obtaining dispatching person's information and order information；
Order probability prediction unit, for being matched using dispatching person's information and order information calculating according to machine learning method
The order probability for the person of sending；
Matching unit carries out global optimization's matching according to the order probability of the dispatching person, to determine order and dispatching person
Matching relationship；
Order is distributed to the matched dispatching person of global optimum according to the matching relationship by order dispatch unit.
6. device according to claim 5, which is characterized in that the order probability prediction unit includes:
Probability calculation subelement, the order that dispatching person is calculated using feature weight vector sum bias in Growth Function are general
Rate；
Feature weight vector iteration subelement updates institute according to the loss function of setting and feature weight vector iterative calculation
State feature weight vector.
7. device according to claim 5, which is characterized in that the matching unit includes:
Assessment unit is matched, based on being weighted according to matching weighted value set in advance to the order probability of the dispatching person
It calculates, obtains dispatching person and match assessed value；
Global optimum's matching primitives unit calculates matching summation according to the matching assessed value of the dispatching person and allocation strategy score value
Value, corresponding allocation strategy is as the allocation strategy of optimization when will match total value highest, according to the allocation strategy of optimization
Determine order and the matching relationship of dispatching person.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811180185.3A CN109508862A (en) | 2018-10-10 | 2018-10-10 | A kind of allocator and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811180185.3A CN109508862A (en) | 2018-10-10 | 2018-10-10 | A kind of allocator and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109508862A true CN109508862A (en) | 2019-03-22 |
Family
ID=65746468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811180185.3A Pending CN109508862A (en) | 2018-10-10 | 2018-10-10 | A kind of allocator and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109508862A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020827A (en) * | 2019-04-17 | 2019-07-16 | 重庆淘创科技有限公司 | A kind of intelligent distribution method |
| CN111310119A (en) * | 2020-02-10 | 2020-06-19 | 拉扎斯网络科技（上海）有限公司 | Distribution method, distribution device, server and storage medium of distribution tasks |
| CN112836914A (en) * | 2019-11-25 | 2021-05-25 | 北京三快在线科技有限公司 | Order structure adjusting method and device, storage medium and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002037234A2 (en) * | 2000-10-20 | 2002-05-10 | Brown Kenneth B | System and method for collaborative order fulfillment |
| CN105117777A (en) * | 2015-07-28 | 2015-12-02 | 北京嘀嘀无限科技发展有限公司 | Order distributing method and apparatus |
| CN106971279A (en) * | 2017-05-03 | 2017-07-21 | 百度在线网络技术（北京）有限公司 | Estimate method, device, equipment and the storage medium of driver's order behavior |
| CN107203789A (en) * | 2017-06-23 | 2017-09-26 | 北京小度信息科技有限公司 | Distribution model method for building up, distribution method and relevant apparatus |
| CN108229864A (en) * | 2018-03-05 | 2018-06-29 | 北京三快在线科技有限公司 | Distribution method, device and the electronic equipment of order |
| CN108537352A (en) * | 2017-03-02 | 2018-09-14 | 北京嘀嘀无限科技发展有限公司 | A kind of data processing method, device and server |
-
2018
- 2018-10-10 CN CN201811180185.3A patent/CN109508862A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002037234A2 (en) * | 2000-10-20 | 2002-05-10 | Brown Kenneth B | System and method for collaborative order fulfillment |
| CN105117777A (en) * | 2015-07-28 | 2015-12-02 | 北京嘀嘀无限科技发展有限公司 | Order distributing method and apparatus |
| CN108537352A (en) * | 2017-03-02 | 2018-09-14 | 北京嘀嘀无限科技发展有限公司 | A kind of data processing method, device and server |
| CN106971279A (en) * | 2017-05-03 | 2017-07-21 | 百度在线网络技术（北京）有限公司 | Estimate method, device, equipment and the storage medium of driver's order behavior |
| CN107203789A (en) * | 2017-06-23 | 2017-09-26 | 北京小度信息科技有限公司 | Distribution model method for building up, distribution method and relevant apparatus |
| CN108229864A (en) * | 2018-03-05 | 2018-06-29 | 北京三快在线科技有限公司 | Distribution method, device and the electronic equipment of order |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020827A (en) * | 2019-04-17 | 2019-07-16 | 重庆淘创科技有限公司 | A kind of intelligent distribution method |
| CN112836914A (en) * | 2019-11-25 | 2021-05-25 | 北京三快在线科技有限公司 | Order structure adjusting method and device, storage medium and electronic equipment |
| CN111310119A (en) * | 2020-02-10 | 2020-06-19 | 拉扎斯网络科技（上海）有限公司 | Distribution method, distribution device, server and storage medium of distribution tasks |
| CN111310119B (en) * | 2020-02-10 | 2020-11-17 | 拉扎斯网络科技（上海）有限公司 | Distribution method, distribution device, server and storage medium of distribution tasks |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kahraman et al. | Hospital location selection using spherical fuzzy TOPSIS | |
| Şahin et al. | A Multi-criteria neutrosophic group decision making metod based TOPSIS for supplier selection | |
| Cho | Tourism forecasting and its relationship with leading economic indicators | |
| Tsai et al. | A novel model of technology diffusion: System dynamics perspective for cloud computing | |
| CN109766454A (en) | A kind of investor's classification method, device, equipment and medium | |
| Dyatlov et al. | Information space convergence as a new stage of e-governance development in Eurasian economic space | |
| CN109508862A (en) | A kind of allocator and device | |
| Mandic et al. | Two-phase model for multi-criteria project ranking: Serbian Railways case study | |
| US20140324528A1 (en) | Computerized System and Method for Determining an Action's Relevance to a Transaction | |
| CN101963983A (en) | Data mining method of rough set and optimization neural network | |
| Lin et al. | Tourism demand forecasting: Econometric model based on multivariate adaptive regression splines, artificial neural network and support vector regression | |
| Vafadarnikjoo et al. | A hybrid approach of intuitionistic fuzzy set theory and DEMATEL method to prioritize selection criteria of bank branches locations | |
| Sopha et al. | Location planning of urban distribution center under uncertainty: A case study of Yogyakarta special region province, Indonesia | |
| CN112150291A (en) | Intelligent financial product recommendation system | |
| Rogers et al. | The inclusion of non-market values in systematic conservation planning to enhance policy relevance | |
| Abel et al. | The form and evolution of international migration networks, 1990–2015 | |
| Lin et al. | Currency exchange rates prediction based on linear regression analysis using cloud computing | |
| CN106203864A (en) | A kind of brand assets appraisal procedure based on big data and system | |
| Micus et al. | Methods to analyze customer usage data in a product decision process: A systematic literature review | |
| CN113052653A (en) | Financial product content recommendation method and system and computer readable storage medium | |
| Quynh | An Integrated Dynamic Generalized Trapezoidal Fuzzy AHPTOPSIS Approach for Evaluating Sustainable Performance of Bank | |
| CN116629904A (en) | Client hierarchical matching method based on big data | |
| Gholamveisy et al. | A Hybrid GRA-VIKOR Approach for Prioritization of Sales Management Outsourcing Risks (Case: energy company) | |
| Agagu et al. | Context-aware recommendation methods | |
| Cho et al. | A robust time series prediction model using POMDP and data analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190322 |
|
| RJ01 | Rejection of invention patent application after publication |