CN108122042A - Distribution time predictor method and device - Google Patents
Distribution time predictor method and device Download PDFInfo
- Publication number
- CN108122042A CN108122042A CN201611069127.4A CN201611069127A CN108122042A CN 108122042 A CN108122042 A CN 108122042A CN 201611069127 A CN201611069127 A CN 201611069127A CN 108122042 A CN108122042 A CN 108122042A
- Authority
- CN
- China
- Prior art keywords
- order
- time
- duration
- section
- execution
- 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
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000009826 distribution Methods 0.000 title claims abstract description 19
- 230000009471 action Effects 0.000 claims abstract description 188
- 238000012545 processing Methods 0.000 claims abstract description 42
- 230000000694 effects Effects 0.000 claims description 40
- 238000000605 extraction Methods 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000010801 machine learning Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 8
- 230000001186 cumulative effect Effects 0.000 claims description 5
- 230000000875 corresponding effect Effects 0.000 description 47
- 230000008569 process Effects 0.000 description 33
- 238000010586 diagram Methods 0.000 description 15
- 238000004590 computer program Methods 0.000 description 9
- 238000003860 storage Methods 0.000 description 9
- 239000000284 extract Substances 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000012163 sequencing technique Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013410 fast food Nutrition 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present application provides a kind of distribution time predictor method and device.Distribution time predictor method includes:It determines pending order, there are multiple action sections in the processing stream of pending order;Estimate multiple execution durations corresponding to multiple action sections;According to multiple execution durations, the estimated delivery time of pending order is obtained.Using the embodiment of the present application, the accuracy of the estimated delivery time of order can be improved, reduces actual service time and it is expected that the error between delivery time of order.
Description
Technical field
This application involves Internet technical field more particularly to a kind of distribution time predictor methods and device.
Background technology
With the fast development of Internet technology, the application based on internet is more and more, such as takes out class application, shopping
Class application.Based on these applications, user stays indoors the article that can be obtained needed for oneself.These are applied in the same of convenient user
When, item dispenser problem is also faced with, then logistic dispatching system comes into being.
Logistic dispatching system not only needs to solve Order Allocation, it is also necessary to provide the estimated delivery time of order.Mesh
Before, logistic dispatching system can assume lower single time dispatching order of the dispatching person according to order, based on dispatching order, calculate and order
Single dispatching distance, and the average speed of dispatching person is combined, obtain the estimated delivery time of order.
The content of the invention
It is found after a large amount of orders of inventor's follow-up study, actual service time of many orders and it is expected that delivery time error
It is larger, it is contemplated that delivery time is not accurate enough.
In view of the above-mentioned problems, inventor has re-started research to the scheme of existing estimated delivery time, find:Existing side
Case directly counts the dispatching distance of order, and with reference to the average speed of dispatching person, the ratio for obtaining distance and speed is sent as estimated
Up to the time, this mode is simple, efficient, but granularity is thicker, fails to fully take into account the complexity of order delivery process, so
It is expected that delivery time is not accurate enough.
Based on above-mentioned, present inventor proposes a solution, and cardinal principle is:According in order processing flow
Action, multiple action sections are split as by the processing procedure of order, estimate the execution duration of each action section, then comprehensive multiple
The execution duration of section is acted, obtains the estimated delivery time of order.Granularity is estimated by reducing, improves estimated delivery time
Accuracy reduces actual service time and it is expected that the error between delivery time of order.
Based on above-mentioned, the embodiment of the present application provides a kind of distribution time predictor method, including:
It determines pending order, there are multiple action sections in the processing stream of the pending order;
Estimate multiple execution durations corresponding to the multiple action section;
According to the multiple execution duration, the estimated delivery time of the acquisition pending order.
In an optional embodiment, the acquisition step of the estimated delivery time, including:According to the pending order
Tapeout state, determine to estimate initial time;Initial time is estimated based on described, add up the multiple execution duration, to obtain
The estimated delivery time.
In an optional embodiment, before cumulative the multiple execution duration, the method further includes:According to described
Estimate initial time and the multiple execution duration, calculate the multiple action section it is corresponding perform the time started and/
Or perform the end time;According to it is the multiple action section it is corresponding perform the time started and/or perform the end time it
Between sequencing, correct the multiple execution duration.
In an optional embodiment, the method further includes:Export the multiple action corresponding execution of section
Time started and/or execution end time.
In an optional embodiment, the multiple execution time estimates step, including:Determine the multiple active region
Section is corresponding to estimate logic;Performing the multiple action, section is corresponding estimates logic, to obtain the multiple execution duration.
In an optional embodiment, the multiple action section correspondence estimates the execution step of logic, including:From described
In the pending associated multi-dimensional data of order, the multiple action section each associated dimensional characteristics are extracted;According to described
It is multiple to act sections each associated dimensional characteristics, generate the multiple action respective action parameter of section;Based on described more
A action respective action parameter of section, obtains the multiple execution duration.
In an optional embodiment, the multiple action section includes:To shop section;
The extraction step to the associated dimensional characteristics of shop section, including:From the multi-dimensional data, order is extracted
Position feature and the speed effect characteristics of dispatching person, wherein, the order position feature is the pending affiliated order of order
The position of each order in group;
The generation step of the action parameter to shop section, including:According to the order position attribution, planning takes single channel
Line;According to the speed effect characteristics and First Speed model, calculating takes one velocity;
The acquisition step of the execution duration to shop section, including:According to it is described take single channel line and it is described take one velocity,
Get shop duration.
In an optional embodiment, the method further includes:From the History Order data of the dispatching person, acquisition is gone through
History speed effect characteristics and history dispatching duration;Using the historical speed effect characteristics and history dispatching duration as training
Sample trains the First Speed model using machine learning algorithm.
In an optional embodiment, the multiple action section includes:Take single section;
The extraction step for taking the associated dimensional characteristics of single section, including:From the multi-dimensional data, described in extraction
The order detail feature of pending order and trade company's attributive character;
The generation step of the action parameter for taking single section, including:Belonged to according to the order detail feature and trade company
Property feature, with reference to single model is gone out, generates single duration;
The acquisition step for performing duration for taking single section, including:According to lower single time of the pending order and
It is described go out single duration, determine single time;Estimate the dispatching person of the pending order to the shop time;According to it is described to shop when
Between and it is described go out single time, acquisition take single duration.
In an optional embodiment, the acquisition step for taking single duration, including:If described arrive the shop time earlier than described
Go out single time, the time difference between single time and the time to shop is gone out according to compensating incremental time, to be taken described in acquisition
Single duration;If it is described to the shop time be later than it is described go out single time, the incremental time is taken into single duration as described in.
In an optional embodiment, the method further includes:From the History Order number of the trade company of the pending order
In, extraction History Order details feature, history trade company attributive character and history go out single duration;History Order details are special
Sign, history trade company attributive character and history go out single duration as training sample, using machine learning algorithm train described in go out list
Model.
In an optional embodiment, the multiple action section includes:To user hotline;
The extraction step to the associated dimensional characteristics of user hotline, including:From the multi-dimensional data, extraction is ordered
Single position feature and the speed effect characteristics of dispatching person;
The generation step of the action parameter to user hotline, including:According to the order position attribution, list is sent in planning
Route;According to the speed effect characteristics and second speed model, one velocity is sent in calculating;
The acquisition step of the execution duration to user hotline, including:It send single channel line according to described and described send single speed
Degree, gets user's duration.
In an optional embodiment, the multiple action section includes:Etc. user hotlines;
The extraction step of the associated dimensional characteristics of user hotlines such as described, including:From the multi-dimensional data, acquisition is matched somebody with somebody
The History Order data for the person of sending;From the History Order data, the historical position and History Order of extracting the dispatching person are used
The band of position at family;
The generation step of the action parameter of the user hotlines such as described, including:According to the historical position of the dispatching person and institute
The band of position of History Order user is stated, the dispatching person is counted and is located at being averaged in the band of position of the History Order user
Duration;
The acquisition step of the execution duration of the user hotlines such as described, including:According to the average duration, obtain as users
It is long.
Correspondingly, the embodiment of the present application also provides a kind of distribution time estimating device, including:
Determination unit for determining pending order, has multiple action sections in the processing stream of the pending order;
Unit is estimated, for estimating multiple execution durations corresponding to the multiple action section;
Time processing unit, for according to the multiple execution duration, obtaining when being expected to be sent to of the pending order
Between.
In an optional embodiment, the time processing unit is specifically used for:According to placing an order for the pending order
State determines to estimate initial time;Initial time is estimated based on described, add up the multiple execution duration, described pre- to obtain
Count delivery time.
In an optional embodiment, the time processing unit is additionally operable to:Before cumulative the multiple execution duration,
Initial time and the multiple execution duration are estimated according to described, the multiple corresponding execution of action section is calculated and starts
Time and/or execution end time;Terminated according to the multiple action section corresponding execution time started and/or execution
Sequencing between time corrects the multiple execution duration.
In an optional embodiment, the time processing unit is additionally operable to:It is each right to export the multiple action section
The execution time started and/or execution end time answered.
In an optional embodiment, the unit of estimating includes:
Determination subelement, for determining the multiple action, section is corresponding estimates logic;
Subelement is performed, section is corresponding estimates logic for performing the multiple action, to obtain the multiple execution
Duration.
In an optional embodiment, the execution subelement includes:
Extracting sub-module, for from the pending associated multi-dimensional data of order, extracting the multiple active region
The respective associated dimensional characteristics of section;
Submodule is generated, for each associated dimensional characteristics, generation to be the multiple dynamic according to the multiple action section
Make the respective action parameter of section;
Acquisition submodule, for being based on the multiple action respective action parameter of section, when obtaining the multiple execution
It is long.
In an optional embodiment, the multiple action section includes:To shop section;
The extracting sub-module is specifically used for:From the multi-dimensional data, order position feature and dispatching person are extracted
Speed effect characteristics, wherein, the order position feature is the position of each order in the pending affiliated order group of order;
The generation submodule is specifically used for:According to the order position attribution, planning takes single channel line;According to the speed
Effect characteristics and First Speed model, calculating take one velocity;
The acquisition submodule is specifically used for:According to it is described take single channel line and it is described take one velocity, get shop duration.
In an optional embodiment, the multiple action section includes:Take single section;
The extracting sub-module is specifically used for:From the multi-dimensional data, the order for extracting the pending order is detailed
Feelings feature and trade company's attributive character;
The generation submodule is specifically used for:According to the order detail feature and trade company's attributive character, with reference to going out list
Model generates single duration;
The acquisition submodule is specifically used for:According to lower single time of the pending order and it is described go out single duration, really
Make single time;Estimate the dispatching person of the pending order to the shop time;According to it is described to the shop time and it is described go out it is single when
Between, acquisition takes single duration.
In an optional embodiment, the multiple action section includes:To user hotline;
The extracting sub-module is specifically used for:From the multi-dimensional data, order position feature and dispatching person are extracted
Speed effect characteristics;
The generation submodule is specifically used for:According to the order position attribution, single channel line is sent in planning;According to the speed
One velocity is sent in effect characteristics and second speed model, calculating;
The acquisition submodule is specifically used for:According to it is described send single channel line and it is described send one velocity, get user's duration.
In an optional embodiment, the multiple action section includes:Etc. user hotlines;
The extracting sub-module is specifically used for:From the multi-dimensional data, the History Order data of dispatching person are obtained;From
In the History Order data, the historical position of the dispatching person and the band of position of History Order user are extracted
The generation submodule is specifically used for:According to the historical position of the dispatching person and the position of the History Order user
Region is put, counts the average duration that the dispatching person is located in the band of position of the History Order user;
The acquisition submodule is specifically used for:According to users' durations such as the average duration, acquisitions.
In the embodiment of the present application, the processing procedure of order is divided into multiple action sections, estimates each action section
Execution duration, then execution durations of comprehensive multiple action sections obtain the estimated delivery time of order.Due to estimating granularity phase
To relatively thin, sectional is estimated, and can reduce each action section estimates influencing each other between effect;Additionally, it is contemplated that active region
Relevance between section, the predictor error of certain action section it is also possible to obtain centainly in the process in estimating for relevant action section
The delivery time of order is estimated in the compensation of degree, therefore, sectional, is conducive to improve the accuracy of estimated delivery time, is reduced
The actual service time of order and it is expected that error between delivery time.
Description of the drawings
Attached drawing described herein is used for providing further understanding of the present application, forms the part of the application, this Shen
Schematic description and description please does not form the improper restriction to the application for explaining the application.In the accompanying drawings:
Fig. 1 is the flow diagram for the distribution time predictor method that one embodiment of the application provides;
Fig. 2 is the flow diagram for performing duration estimated to shop section that another embodiment of the application provides;
Fig. 3 is the flow diagram for the pre- execution duration for being estimated to take single section that the another embodiment of the application provides;
Fig. 4 is the flow diagram for the execution duration for estimating user hotline that the another embodiment of the application provides;
Fig. 5 is the flow diagram of the execution duration of the user hotlines such as estimating of providing of the another embodiment of the application;
Fig. 6 is the structure diagram for the distribution time estimating device that the another embodiment of the application provides;
Fig. 7 is the structure diagram for the distribution time estimating device that the another embodiment of the application provides.
Specific embodiment
To make the purpose, technical scheme and advantage of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one
Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Go out all other embodiments obtained under the premise of creative work, shall fall in the protection scope of this application.
Fig. 1 is the flow diagram for the distribution time predictor method that one embodiment of the application provides.It is as shown in Figure 1, described
Method includes:
101st, determine pending order, there are multiple action sections in the processing stream of the pending order.
102nd, multiple execution durations corresponding to multiple action sections are estimated.
103rd, according to multiple execution durations, the estimated delivery time of pending order is obtained.
In the present embodiment, it would be desirable to which the order for providing estimated delivery time is referred to as pending order.Optionally, wait to locate
It can be new order to manage order.In an application example, after logistic dispatching system reception user places an order, not only need for new order
Distribute dispatching person, it is also necessary to provide the estimated delivery time of new order, and estimated delivery time is shown in the order of user terminal
On details page, situation is sent to so that user understands order, improves user experience.In this regard, this may be employed in logistic dispatching system
The method that embodiment provides provides the estimated delivery time of new order.
In the present embodiment, the relevant action in order dealing process carries out section partition to order dealing process.
For pending order, after section partition, there are multiple action sections in processing stream.
In the present embodiment, the action form of division action section institute foundation is not intended to limit, on every order processing stream
Action can be as the node that division action section is used in the present embodiment.In addition, there are multiple actions, no on order processing stream
Certain each action will be used for division action section, can be made according to application demand, all adaptability selected section or action
For partitioning site, so that order processing stream is divided into multiple action sections.
For example, the action on order processing stream includes but not limited to:System order, system distribute, dispatching person dispenses,
Notifications trade company, trade company's processing etc..Wherein, cited partial act can be segmented further, be acted with dispatching person's dispatching
Exemplified by, and can be subdivided into:Trade company, dispatching person belonging to dispatching person's arrival order take list, dispatching person to reach the use such as user, dispatching person
Family, user such as sign at the actions.In the present embodiment, can be from the above-mentioned action enumerated, selected section or all action conduct
The processing stream of pending order is divided into multiple action sections by partitioning site.
For example, system order, system distribution and dispatching person, which may be employed, dispenses processing of three actions to pending order
Stream is divided, then the processing stream of pending order includes:Order section, distribution section and dispatching section.Wherein, order section
Refer to the process of the success order from the initial point of setting to system;Distribution section refers to will to system since system success order
Order successfully distributes to the process of dispatching person;Dispatching section refers to since order is successfully distributed to dispatching person by system to dispatching
Order is successfully sent to the process of user by member.
In another example the action in dispatching person's delivery process may be employed, for example, dispatching person take list, dispatching person reach user and
The users such as dispatching person divide the processing stream of pending order, then the processing stream of pending order includes:Take order area
Section send order section and waits user hotlines.Wherein, order section is taken to refer to since the initial point of setting, to dispatching person from treating
Handle the process that success at the affiliated trade company of order obtains order;Optionally, the initial point of setting can be system success order.It send
Order section refers to since dispatching person's success obtains order, and the process of user region is reached to dispatching person;Wait user areas
Section refers to since dispatching person reaches user region, pending order successfully be signed for user or dispatching person leaves user institute
Process in region.
What deserves to be explained is above-mentioned name and definition to each action section, is had according to what generic scenario provided
The name and definition of universality, however it is not limited to this, can be provided according to concrete application scene other definition or to it is above-mentioned define into
Row adaptation.
In the present embodiment, the dispatching distance of order is directly counted unlike the prior art, with reference to the flat of dispatching person
Equal speed obtains distance and estimated delivery time of the ratio of speed as pending order, but based on to pending order
Processing stream on multiple action sections, estimate the corresponding multiple execution durations of multiple action sections, and according to multiple active regions
The corresponding multiple execution durations of section obtain the estimated delivery time of pending order.Wherein, multiple action sections and multiple execution
Duration corresponds.The execution duration of action section refers to complete the time span needed for the action section.
In the present embodiment, the delivery time of order is estimated using the action section on order processing stream as granularity, estimates grain
Spend relatively thin, sectional is estimated, and can reduce each action section estimates influencing each other between effect;It is additionally, it is contemplated that dynamic
Make the relevance between section, the predictor error of certain action section it is also possible to obtain in the process in estimating for relevant action section
The delivery time of order is estimated in a degree of compensation, therefore, sectional, is conducive to improve the accuracy of estimated delivery time.
For example, in an application scenarios, estimated delivery time that logistic dispatching system can be provided based on the present embodiment is
Order or the dispatching person's marking for being responsible for dispatching order.Since the estimated delivery time that the present embodiment provides is more accurate,
Logistic dispatching system can be more accurately the order or the dispatching person's marking for being responsible for dispensing the order, to improve
Logistic Scheduling logic promotes service quality.
In another example in an application scenarios, logistic dispatching system can be inputted to user terminal the present embodiment provide it is estimated
Delivery time.Since the estimated delivery time that the present embodiment provides is more accurate, order is more accurately understood convenient for user and is sent
The time range reached is conducive to improve user experience.
In above-described embodiment or following embodiments, estimate it is multiple action sections it is corresponding it is multiple execution duration the step of,
Can be:Determine that section is corresponding estimates logic for multiple actions;Performing multiple actions, section is corresponding estimates logic, more to obtain
A execution duration.
In an application example, using multiple action sections to be whole, for multiple action sections correspondence is set to estimate logic, and
It establishes multiple action sections and estimates the mapping relations between logic.In the application example, multiple action sections correspond to one
Perform logic.Based on this, according to multiple action sections and the mapping relations between logic can be estimated, determine multiple action sections
It is corresponding to estimate logic.When estimating logic described in execution, multiple action respective execution durations of section can be obtained.
In an application example, can be multiple action sections be correspondingly arranged it is multiple estimate logic, and establish multiple actions
Section and multiple mapping relations estimated between logic.In the application example, it is multiple action sections with it is multiple estimate logic it
Between correspond.Based on this, can be determined multiple dynamic according to multiple action sections and multiple mapping relations estimated between logic
Make that section is corresponding to estimate logic.When perform it is multiple action section is corresponding estimate logic when, can obtain multiple
Act the respective execution duration of section.
The corresponding logic of estimating of multiple action sections is mainly used for obtaining multiple action respective execution durations of section.According to
The difference of application demand, the corresponding realization for estimating logic of the multiple action section will be different.Wherein, logic is estimated
Quality directly affects the accuracy of the execution duration of corresponding actions section, and therefore, the following embodiments of the application provide a kind of optimization
Estimate logic, with improve perform estimate logic acquisition execution duration accuracy, further improve estimated delivery time
Accuracy.
Preferably, the corresponding logic of estimating of multiple action sections includes:It extracts dimensional characteristics, moved based on dimensional characteristics generation
Make parameter, the operations such as execution duration are calculated based on action parameter.It is multiple to act the corresponding execution for estimating logic of section based on this
Step, Ke Yiwei:From the associated multi-dimensional data of pending order, extracting multiple action sections, each associated dimension is special
Sign;Sections each associated dimensional characteristics are acted according to multiple, generate multiple action respective action parameters of section;Based on multiple
The respective action parameter of section is acted, obtains the corresponding multiple execution durations of multiple action sections.
With reference to specific action section, the above-mentioned implementation procedure for estimating logic is described in detail.
In a kind of application example, according to the action of dispatching person in order delivery process, order dealing process is split as
Following action section:To shop section, take single section, to user hotline and wait user hotlines.Wherein, it is to shop section
Refer to since the initial point of setting, reached to dispatching person at the trade company belonging to pending order;Optionally, the initial point of setting can
Be system success order.Single section is taken to refer to since the trade company belonging to dispatching person reaches pending order, obtain to success
The process of order.Refer to user hotline since dispatching person's success obtains order, user region is reached to dispatching person
Process.Etc. user hotlines refer to since dispatching person reaches user region, pending order successfully be signed for user or is matched somebody with somebody
The person of sending leaves the process of user region.
What deserves to be explained is above-mentioned name and definition to each action section is only a kind of tool provided according to generic scenario
There are the name and definition of universality, however it is not limited to this, other definition can be provided according to concrete application scene or to above-mentioned definition
Carry out adaptation.
To arriving shop section:
The associated dimensional characteristics of shop section can be extracted from the associated multi-dimensional data of pending order;According to
Section associated dimensional characteristics in shop are generated to the action parameter of shop section;According to the action parameter to shop section, shop area is acquired
The corresponding execution duration of section, referred to as to shop duration.Wherein, the idiographic flow of shop duration is got, as shown in Fig. 2, including:
201st, from the associated multi-dimensional data of pending order, extracting the speed of order position feature and dispatching person influences
Feature.
Inventor has found by researching and analysing:Dispatching person, which is primarily referred to as, to shop section reaches the pending affiliated trade company of order
The process at place, for this process, important influence factor is to take single channel line and dispatching person's speed.
Optionally, optimization algorithm and machine learning method can be combined, is cooked up for pending order for dispatching person
One most reasonably takes single channel line.For example, single act can be taken by simulate dispatching person, it is reasonable to be carried out from the angle of dispatching person
Planning so that cooks up takes single channel line to take single channel line close to dispatching person's reality as far as possible, can so reduce and be based on being planned
The accuracy for taking the error that single channel line generates when estimating the delivery time of order, promoting estimated delivery time.
Wherein, single channel line is taken using algorithms of different planning, corresponding effect characteristics simultaneously differ.In the present embodiment
In, by taking minimal path algorithm as an example, then influence to take single channel line is mainly characterized by each order in the affiliated order group of pending order
Position and the current position of dispatching person.Based on this, order position can be extracted from the associated multi-dimensional data of pending order
Put feature.The order position feature includes the position of each order in the affiliated order group of pending order.
Dispatching person's speed is mainly affected by factors, these factors are referred to as speed effect characteristics.Based on this, from treating
It manages in the associated multi-dimensional data of order, extracts the speed effect characteristics of dispatching person.Dispatching person here refers to that being responsible for dispatching treats
Handle the dispatching person of order, the speed effect characteristics of dispatching person can be the factor of the dispatching speed of any influence dispatching person, example
Such as vehicles of the person's of dispatching use, weather conditions, the gender of dispatching person, age.
202nd, according to order position attribution, planning takes single channel line, and according to speed effect characteristics and First Speed model, meter
Calculation takes one velocity.
The order position feature and speed effect characteristics extracted based on step 201, is performed step 202, can be generated to shop area
The action parameter of section.Action parameter to shop section includes taking single channel line and takes one velocity.
In the present embodiment, it is contemplated that the dispatching custom of different dispatching persons itself, style etc. are all different, therefore from dispatching person's
With reference to a variety of speed effect characteristics, a kind of rate pattern is trained for dispatching person for angle, the History Order data based on dispatching person,
Referred to as First Speed model.For example, from the History Order data of dispatching person, historical speed effect characteristics and history are gathered
Dispense duration;Historical speed effect characteristics and history are dispensed into duration as training sample, trained using machine learning algorithm
First Speed model.Rate pattern based on dispatching person itself, and consider a variety of speed effect characteristics, calculate dispatching person's
One velocity is taken, is conducive to improve the accuracy for taking one velocity.
203rd, according to taking single channel line and taking one velocity, shop duration is got.
Based on the action parameter to shop section that step 202 generates, step 203 is performed, the execution of shop section can be acquired
Duration, i.e., to shop duration.For example, it, according to taking single distance and taking one velocity, can be obtained according to the calculating of single channel line is taken to take single distance
To shop duration.
To during the estimating of shop section, the rate pattern of conjunctive path planning and dispatching person are conducive to improve pre-
Estimate the accuracy of shop duration, and then be conducive to improve the accuracy of estimated delivery time.
To taking single section:
The associated dimensional characteristics of single section can be taken from the associated multi-dimensional data of pending order;According to taking
Single associated dimensional characteristics of section, generation take the action parameter of single section;According to the action parameter for taking single section, Qu Dan areas are obtained
The corresponding execution duration of section, referred to as takes single duration.Wherein, the idiographic flow of single duration is taken, as shown in figure 3, including:
301st, from the multi-dimensional data of pending order, the order detail feature and trade company of extracting pending order belong to
Property feature.
Inventor has found by researching and analysing:Single section is taken to be primarily referred to as dispatching person at the affiliated trade company of pending order
Success obtains the process of order, and for this process, important influence factor is that trade company goes out single time.Wherein, on
Stating out single time refers to that trade company completes pending order, and pending order is made to be in the time of desirable condition.
Wherein, trade company go out that single time is mainly reflected in trade company go out single duration, i.e., the above-mentioned pending order of completion needs
The time span of cost.The details and trade company's attributive character for going out single duration according to pending order are related.It sells in addition in scene
Order exemplified by, order detail feature is primarily referred to as the food product type of order point, quantity etc.;Trade company's attributive character is mainly trade company
Classification, such as fast food, Chinese meal, western-style food.
Based on above-mentioned, from the multi-dimensional data of pending order, can extract the order detail feature of pending order with
And trade company's attributive character.
302nd, according to order detail feature and trade company's attributive character, with reference to single model is gone out, single duration is generated.
The order detail feature and trade company's attributive character extracted based on step 301, perform step 302, and list is taken with generation
The action parameter of section.Taking the action parameter of single section includes single duration.
Optionally, it is contemplated that the order identical to order detail, different trade companies to go out single duration generally different;It is detailed to order
The different order of feelings, single duration that goes out of same trade company also can be different.It is therefore possible to use machine learning method, belongs to reference to trade company
The feature of multiple dimensions such as property feature, order detail feature trains the single model that goes out of the affiliated trade company of pending order, so as to pre-
That estimates each order under the trade company goes out single duration.For example, it can be extracted from the History Order data of the trade company of pending order
History Order details feature, history trade company attributive character and history go out single duration;By History Order details feature, history trade company
Attributive character and history go out single duration as training sample, using machine learning algorithm train described in go out single model.
The single model that goes out in the present embodiment considers the multi-dimensional datas such as trade company, order simultaneously, therefore is based on out single model
The accuracy for going out single duration obtained is higher, is conducive to improve the accuracy of estimated delivery time.
303rd, according to lower single time of pending order and go out single duration, determine single time.
For example, lower single time can be based on, add up out single duration, as going out single time.
304th, estimate the dispatching person of pending order to the shop time.
Optionally, the flow that can perform in embodiment similar to Figure 2 is determined to shop duration, with reference to lower single time, is determined
To the shop time.Alternatively, can be directly based upon in embodiment illustrated in fig. 2 to shop duration, with reference to lower single time, when determining to shop
Between.
305th, basis to the shop time and goes out single time, and acquisition takes single duration.
For example, comparable dispatching person's goes out single time to shop time and trade company;If it is said to the shop time earlier than single time is gone out
Bright dispatching person needs to wait, and goes out single time and is exactly to take single duration substantially to the time difference between the time of shop, but considers actual feelings
Dispatching person needs shop, startup dispatching instrument etc. to be also required to the time after taking list in condition, therefore can compensate out list according to incremental time
Time and to the time difference between the time of shop, single duration is taken to obtain;If being later than single time to the shop time, illustrate dispatching person's nothing
It needs to wait for, can directly take list, similarly, consider that dispatching person needs shop after taking list, starts dispatching instrument etc. in actual conditions
The time is needed, therefore can be using incremental time as taking single duration.
In the present embodiment, the not value of limiting time increment, depending on application demand, such as can be 1 point
Clock, 1.5 minutes etc..
It can be seen that going out single duration based on 302 generations, step 303-305 is performed, when acquisition takes the execution of single section
It is long, that is, take single duration.
During the estimating of single section is taken, be based on out single model, be attached to the shop time and go out single time comparison and
The compensation of incremental time is conducive to improve the accuracy for being estimated to take single duration in advance, and then is conducive to improve the standard of estimated delivery time
True property.
To arriving user hotline:
The associated dimensional characteristics of user hotline can be extracted from the associated multi-dimensional data of pending order;According to
To the associated dimensional characteristics of user hotline, the action parameter of user hotline is generated to;According to the action parameter to user hotline, obtain
The corresponding execution duration of user hotline is obtained, referred to as to user's duration.Wherein, the idiographic flow of shop duration, such as Fig. 4 are got
It is shown, including:
401st, from the associated multi-dimensional data of pending order, extracting the speed of order position feature and dispatching person influences
Feature.
Inventor has found by researching and analysing:Dispatching person is primarily referred to as to user hotline and takes to reach the process of user after list,
For this process, important influence factor is to send single channel line and dispatching person's speed.
Optionally, optimization algorithm and machine learning method can be combined, is cooked up for pending order for dispatching person
One is most reasonably sent single channel line.For example, single act can be sent by simulate dispatching person, it is reasonable to be carried out from the angle of dispatching person
Planning so that cooks up send single channel line to send single channel line close to what dispatching person actually walked as far as possible, can be so reduced based on institute
The accuracy sent the error that single channel line generates when estimating the delivery time of order, promote estimated delivery time of planning.
Wherein, single channel line is sent using algorithms of different planning, corresponding effect characteristics simultaneously differ.In the present embodiment
In, by taking minimal path algorithm as an example, then influence to send single channel line is mainly characterized by each order in the affiliated order group of pending order
Position and the current position of dispatching person.Based on this, order position can be submitted from the associated multi-dimensional data of pending order
Put feature.The order position feature includes the position of each order in the affiliated order group of pending order.
Dispatching person's speed is mainly affected by factors, these factors are referred to as speed effect characteristics.Based on this, from treating
It manages in the associated multi-dimensional data of order, submits the speed effect characteristics of dispatching person.Dispatching person here refers to that being responsible for dispatching treats
Handle the dispatching person of order, the speed effect characteristics of dispatching person can be the factor of the dispatching speed of any influence dispatching person, example
Such as vehicles of the person's of dispatching use, weather conditions, the gender of dispatching person, age.
402nd, according to order position attribution, single channel line is sent in planning, and according to speed effect characteristics and second speed model, meter
One velocity is sent in calculation.
The order position feature and speed effect characteristics submitted based on step 401 is performed step 402, can be generated to user
The action parameter of section.Action parameter to user hotline includes sending single channel line and send one velocity.
In the present embodiment, it is contemplated that the dispatching custom of different dispatching persons itself, style etc. are all different, therefore from dispatching person's
With reference to a variety of speed effect characteristics, a kind of rate pattern is trained for dispatching person for angle, the History Order data based on dispatching person,
Referred to as second speed model.For example, from the History Order data of dispatching person, historical speed effect characteristics and history are gathered
Dispense duration;Historical speed effect characteristics and history are dispensed into duration as training sample, trained using machine learning algorithm
Second speed model.Rate pattern based on dispatching person itself, and consider a variety of speed effect characteristics, calculate dispatching person's
One velocity is sent, is conducive to improve the accuracy for sending one velocity.
What deserves to be explained is second speed model and First Speed model can be same models.Alternatively, second speed mould
Type and First Speed model can also be friction speed models, for example, First Speed model can take one velocity model, second
Rate pattern can send one velocity model.
403rd, according to sending single channel line and sending one velocity, obtain and be sent to user's duration.
Based on the action parameter to shop section that step 402 generates, step 403 is performed, holding for user hotline can be acquired
Row duration, i.e., to user's duration.For example, can according to the calculating of single channel line is sent to send single distance, according to sending single distance with sending one velocity,
Acquire user's duration.
To during the estimating of user hotline, the rate pattern of conjunctive path planning and dispatching person are conducive to improve
The accuracy of user's duration is estimated, and then is conducive to improve the accuracy of estimated delivery time.
Peer users section:
Can be from the associated multi-dimensional data of pending order, the associated dimensional characteristics of user hotlines such as extraction;According to
Etc. the associated dimensional characteristics of user hotlines, generation etc. user hotlines action parameter;According to etc. user hotlines action parameter, obtain
The corresponding execution duration of user hotlines must be waited, referred to as waits users' duration.Wherein, the idiographic flow of users' duration such as acquisition, such as
Shown in Fig. 5, including:
501st, from the multi-dimensional data of pending order, the History Order data of dispatching person are obtained, and from History Order
In data, the historical position of dispatching person and the band of position of History Order user are extracted.
Inventor has found by researching and analysing:Etc. user hotlines be primarily referred to as dispatching person reach user the band of position after
The process of user is waited, the factor for influencing this process is more subjective.In this regard, the present embodiment dispenses History Order according to dispatching person
The duration of user is waited in the process, determines current grade users' duration.Wherein, dispatching person waits for use during dispensing History Order
The duration at family, can be by position of the position (referred to as historical position) with History Order user during dispatching person's dispatching History Order
Region is put to determine.History Order user refers to the user of History Order.Then etc. the associated dimensional characteristics of user hotlines are mainly
The historical position of dispatching person and the band of position of History Order user.Based on this, etc. user hotlines associated dimensional characteristics carry
Take step, Ke Yiwei:From the multi-dimensional data of pending order, the History Order data of dispatching person are obtained.
502nd, according to the historical position of dispatching person and the band of position of History Order user, statistics dispatching person orders positioned at history
Average duration in the band of position of single user.
The historical position of dispatching person submitted based on step 501 and the band of position of History Order user perform step
502, it the action parameter of user hotlines such as can generate.Etc. the action parameters of user hotlines be located at History Order user including dispatching person
The band of position in average duration.
Users' duration such as the average duration the 503rd, being located at according to dispatching person in the band of position of History Order user, acquisition.
Based on step 502 generate etc. user hotlines action parameter, perform step 503, the user hotlines such as can obtain
Duration is performed, that is, waits users' duration.Optionally, the mean time that dispatching person can be located in the band of position of History Order user
It grows, as grade users' duration.Alternatively, the mean time that can also be located at dispatching person in the band of position of History Order user is a length of
Basis, using the incremental time compensation average duration, with users' durations such as acquisitions.
Etc. during the estimating of user hotlines, consider that dispatching person dispenses the mean time of History Order process medium user
It is long, be conducive to improve the accuracy of users' durations such as estimating, and then be conducive to improve the accuracy of estimated delivery time.
In above-described embodiment or following embodiments, according to the corresponding multiple execution durations of multiple sections, obtain pending
The step of estimated delivery time of order, Ke Yiwei:According to the tapeout state of pending order, determine to estimate initial time;Base
In estimating initial time, add up multiple execution durations, to obtain the estimated delivery time of pending order.
Optionally, before cumulative multiple execution durations, can be calculated according to initial time and multiple execution durations is estimated
Multiple action sections are corresponding to be performed the time started and/or performs the end time;It is each corresponded to according to multiple action sections
Perform the time started and/or perform the end time between sequencing, correct multiple execution durations.Based on this, based on pre-
Estimate initial time, add up multiple revised execution durations, to obtain the estimated delivery time of pending order.It is more by correcting
A execution duration can correct the unreasonable situation in segmentation estimation results, reduce predictor error, improve the standard of estimated delivery time
True property.
Optionally, multiple action sections corresponding execution time started can be exported and/or perform the end time.Example
Such as, can be output to the terminal of trade company dispatching person in the section of shop to the shop time, trade company can see dispatching person to shop when
Between, convenient for single order is adjusted flexibly out or accelerates one velocity etc., be conducive to be promoted the Product Experience degree of trade company.It in another example can
To go out single time to the terminal of dispatching person output trade company, what dispatching person can see trade company goes out single time, in order to the person of dispatching
Dispatching speed is adjusted flexibly, is conducive to improve the Product Experience degree of dispatching person.
It should be noted that the executive agent of each step of above-described embodiment institute providing method may each be same equipment,
Alternatively, this method is also by distinct device as executive agent.For example, the executive agent of step 101 to step 103 can be equipment
A;For another example, step 101 and 102 executive agent can be device A, and the executive agent of step 103 can be equipment B;Etc..
Fig. 6 is the structure diagram for the distribution time estimating device that the another embodiment of the application provides.As shown in fig. 6, institute
Showing device includes:Determination unit 61 estimates unit 62 and time processing unit 63.
Determination unit 61 for determining pending order, has multiple action sections in the processing stream of pending order;
Unit 62 is estimated, for estimating multiple execution durations corresponding to multiple action sections;
Time processing unit 63, for according to multiple execution durations, obtaining the estimated delivery time of pending order.
In an optional embodiment, time processing unit 63 is specifically used for:According to the tapeout state of pending order, really
Surely initial time is estimated;Based on initial time is estimated, add up multiple execution durations, to obtain estimated delivery time.
Optionally, time processing unit 63 is additionally operable to before cumulative multiple execution durations:According to estimate initial time and
Multiple execution durations calculate multiple action sections corresponding execution time started and/or perform the end time;According to multiple
The corresponding sequencing performed between time started and/or execution end time of section is acted, when correcting multiple execution
It is long.Correspondingly, time processing unit 63 is particularly used in:Based on initial time is estimated, add up multiple revised execution durations,
To obtain estimated delivery time.
Optionally, time processing unit 63 is additionally operable to:Export it is multiple action sections it is corresponding perform the time starteds and/
Or perform the end time.
In an optional embodiment, as shown in fig. 7, estimating a kind of realization structure of unit 62 includes:Determination subelement
621 and perform subelement 622.
Determination subelement 621, for determining multiple actions, section is corresponding estimates logic.
Subelement 622 is performed, section is corresponding estimates logic for performing multiple actions, to obtain multiple execution durations.
As shown in fig. 7, performing a kind of realization structure of subelement 622 includes:Extracting sub-module 6221, generation submodule
6222 and acquisition submodule 6223.
Extracting sub-module 6221, it is each for from the associated multi-dimensional data of pending order, extracting multiple action sections
The dimensional characteristics of auto correlation;
Submodule 6222 is generated, for acting sections each associated dimensional characteristics according to multiple, generates multiple active regions
The respective action parameter of section;
Acquisition submodule 6223 for being based on multiple action respective action parameters of section, obtains multiple execution durations.
In a kind of application example, according to the action of dispatching person in order delivery process, order dealing process is split as
Following action section:To shop section, take single section, to user hotline and wait user hotlines.Name on each section
Embodiment of the method is can be found in definition, details are not described herein.
To arriving shop section:
Extracting sub-module 6221 is used for:From the associated multi-dimensional data of pending order, it is associated to extract shop section
Dimensional characteristics.Optionally, extracting sub-module 6221 is specifically used for:From the associated multi-dimensional data of pending order, extraction is ordered
Single position feature and the speed effect characteristics of dispatching person.
Generation submodule 6222 is used for:According to the action parameter for the associated dimensional characteristics of shop section, being generated to shop section.
Optionally, generation submodule 6222 is specifically used for:According to order position attribution, planning takes single channel line, and influences spy according to speed
It seeks peace First Speed model, calculating takes one velocity.
Acquisition submodule 6223 is used for:According to the action parameter to shop section, the corresponding execution duration of shop section is acquired,
Referred to as arrive shop duration.Optionally, acquisition submodule 6223 is specifically used for:According to taking single channel line and taking one velocity, shop is got
Duration.
To taking single section:
Extracting sub-module 6221 is used for:From the associated multi-dimensional data of pending order, take single section associated
Dimensional characteristics.Optionally, extracting sub-module 6221 is specifically used for:From the associated multi-dimensional data of pending order, extraction is treated
Handle the order detail feature of order and trade company's attributive character.
Generation submodule 6222 is used for:According to the associated dimensional characteristics of single section are taken, generation takes the action parameter of single section.
Optionally, generation submodule 6222 is specifically used for:According to order detail feature and trade company's attributive character, with reference to going out single model,
Generate single duration.
Acquisition submodule 6223 is used for:According to the action parameter for taking single section, acquisition takes the corresponding execution duration of single section,
Referred to as arrive shop duration.Optionally, acquisition submodule 6223 is specifically used for:When according to lower single time of pending order and going out single
It is long, determine single time;Estimate the dispatching person of pending order to the shop time;According to the shop time and going out single time, obtain
Take single duration.
For example, acquisition submodule 6223 may compare dispatching person go out single time to shop time and trade company;If to the shop time
Earlier than single time is gone out, single time is compensated out and to the time difference between the time of shop according to incremental time, single duration is taken to obtain;If
It is later than single time to the shop time, using incremental time as taking single duration.
To arriving user hotline:
Extracting sub-module 6221 is used for:From the associated multi-dimensional data of pending order, user hotline association is extracted
Dimensional characteristics.Optionally, extracting sub-module 6221 is specifically used for:From the associated multi-dimensional data of pending order, extraction
Order position feature and the speed effect characteristics of dispatching person.
Generation submodule 6222 is used for:According to the associated dimensional characteristics of user hotline, the action of user hotline is generated to
Parameter.Optionally, generation submodule 6222 is specifically used for:According to order position attribution, single channel line is sent in planning, and according to speed shadow
Feature and second speed model are rung, one velocity is sent in calculating.
Acquisition submodule 6223 is used for:According to the action parameter to user hotline, the corresponding execution of user hotline is acquired
Duration, referred to as to user's duration.Optionally, acquisition submodule 6223 is specifically used for:According to sending single channel line and sending one velocity, obtain
It is sent to shop duration.
Peer users section:
Extracting sub-module 6221 is used for:From the associated multi-dimensional data of pending order, the user hotlines association such as extraction
Dimensional characteristics.Optionally, extracting sub-module 6221 is specifically used for:From the multi-dimensional data of pending order, dispatching is obtained
The History Order data of member, and from History Order data, extract the historical position of dispatching person and the position of History Order user
Region.
Generation submodule 6222 is used for:According to etc. the associated dimensional characteristics of user hotlines, generation etc. user hotlines action
Parameter.Optionally, generation submodule 6222 is specifically used for:According to the historical position of dispatching person and the position area of History Order user
Domain, statistics dispatching person are located at the average duration in the band of position of History Order user.
Acquisition submodule 6223 is used for:According to etc. user hotlines action parameter, obtain etc. the corresponding execution of user hotlines
Duration referred to as waits users' duration.Optionally, acquisition submodule 6223 is specifically used for:It is located at History Order according to dispatching person to use
Users' duration such as the average duration in the band of position at family, acquisition.
Distribution time estimating device provided in this embodiment can perform the corresponding flow in above method embodiment, herein
It repeats no more.
The processing procedure of order is divided into multiple action sections, in advance by distribution time estimating device provided in this embodiment
Estimate the execution duration of each action section, then the execution duration of comprehensive multiple action sections, obtain the estimated delivery time of order.
Relatively thin due to estimating granularity, sectional is estimated, and can reduce each action section estimates influencing each other between effect;Separately
Outside, the relevance between consideration action section, certain action predictor error of section is it is also possible to estimating in relevant action section
A degree of compensation is obtained in the process, and therefore, sectional estimates the delivery time of order, is conducive to improve estimated delivery time
Accuracy, reduce actual service time and it is expected that the error between delivery time of order.
It should be understood by those skilled in the art that, the embodiment of the present invention can be provided as method, system or computer program
Product.Therefore, the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware can be used in the present invention
Apply the form of example.Moreover, the computer for wherein including computer usable program code in one or more can be used in the present invention
The computer program production that usable storage medium is implemented on (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The present invention be with reference to according to the method for the embodiment of the present invention, the flow of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that it can be realized by computer program instructions every first-class in flowchart and/or the block diagram
The combination of flow and/or box in journey and/or box and flowchart and/or the block diagram.These computer programs can be provided
The processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that the instruction performed by computer or the processor of other programmable data processing devices is generated for real
The device for the function of being specified in present one flow of flow chart or one box of multiple flows and/or block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction generation being stored in the computer-readable memory includes referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or
The function of being specified in multiple boxes.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that counted
Series of operation steps is performed on calculation machine or other programmable devices to generate computer implemented processing, so as in computer or
The instruction offer performed on other programmable devices is used to implement in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only memory (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic tape cassette, the storage of tape magnetic rigid disk or other magnetic storage apparatus
Or any other non-transmission medium, the information that can be accessed by a computing device available for storage.It defines, calculates according to herein
Machine readable medium does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
It should also be noted that, term " comprising ", "comprising" or its any other variant are intended to nonexcludability
Comprising so that process, method, commodity or equipment including a series of elements are not only including those elements, but also wrap
Include other elements that are not explicitly listed or further include for this process, method, commodity or equipment it is intrinsic will
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that wanted including described
Also there are other identical elements in the process of element, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or the embodiment in terms of combining software and hardware can be used in the application
Form.It is deposited moreover, the application can be used to can use in one or more computers for wherein including computer usable program code
The shape for the computer program product that storage media is implemented on (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The foregoing is merely embodiments herein, are not limited to the application.For those skilled in the art
For, the application can have various modifications and variations.All any modifications made within spirit herein and principle are equal
Replace, improve etc., it should be included within the scope of claims hereof.
Claims (20)
1. a kind of distribution time predictor method, which is characterized in that including:
It determines pending order, there are multiple action sections in the processing stream of the pending order;
Estimate multiple execution durations corresponding to the multiple action section;
According to the multiple execution duration, the estimated delivery time of the acquisition pending order.
2. according to the method described in claim 1, it is characterized in that, the acquisition step of the estimated delivery time, including:
According to the tapeout state of the pending order, determine to estimate initial time;
Initial time is estimated based on described, add up the multiple execution duration, to obtain the estimated delivery time.
3. according to the method described in claim 2, it is characterized in that, before cumulative the multiple execution duration, the method
It further includes:
Initial time and the multiple execution duration are estimated according to described, calculates the multiple action corresponding execution of section
Time started and/or execution end time;
It is suitable according to the priority between the multiple action section corresponding execution time started and/or execution end time
Sequence corrects the multiple execution duration.
4. it according to the method described in claim 3, it is characterized in that, further includes:
It exports the multiple action section corresponding execution time started and/or performs the end time.
5. according to claim 1-4 any one of them methods, which is characterized in that the multiple execution time estimates step,
Including:
Determine that section is corresponding estimates logic for the multiple action;
Performing the multiple action, section is corresponding estimates logic, to obtain the multiple execution duration.
6. according to the method described in claim 5, it is characterized in that, the multiple action section correspondence estimate logic perform step
Suddenly, including:
From the pending associated multi-dimensional data of order, extracting the multiple action section, each associated dimension is special
Sign;
According to the multiple action section, each associated dimensional characteristics, the respective action of the multiple action section of generation are joined
Number;
Based on the multiple action respective action parameter of section, the multiple execution duration is obtained.
7. according to the method described in claim 6, it is characterized in that, the multiple action section includes:To shop section;
The extraction step to the associated dimensional characteristics of shop section, including:
From the multi-dimensional data, order position feature and the speed effect characteristics of dispatching person are extracted, wherein, the order position
Put the position for being characterized as each order in the affiliated order group of the pending order;
The generation step of the action parameter to shop section, including:
According to the order position attribution, planning takes single channel line;
According to the speed effect characteristics and First Speed model, calculating takes one velocity;
The acquisition step of the execution duration to shop section, including:
According to it is described take single channel line and it is described take one velocity, get shop duration.
8. it the method according to the description of claim 7 is characterized in that further includes:
From the History Order data of the dispatching person, historical speed effect characteristics and history dispatching duration are gathered;
Using the historical speed effect characteristics and history dispatching duration as training sample, institute is trained using machine learning algorithm
State First Speed model.
9. according to the method described in claim 6, it is characterized in that, the multiple action section includes:Take single section;
The extraction step for taking the associated dimensional characteristics of single section, including:
From the multi-dimensional data, the order detail feature and trade company's attributive character of the pending order are extracted;
The generation step of the action parameter for taking single section, including:
According to the order detail feature and trade company's attributive character, with reference to single model is gone out, single duration is generated;
The acquisition step for performing duration for taking single section, including:
According to lower single time of the pending order and it is described go out single duration, determine single time;
Estimate the dispatching person of the pending order to the shop time;
According to it is described to the shop time and it is described go out single time, acquisition take single duration.
10. according to the method described in claim 9, it is characterized in that, the acquisition step for taking single duration, including:
If it is described to the shop time earlier than it is described go out single time, go out according to compensating incremental time single time with it is described to the shop time
Between time difference, to take single duration described in acquisition;
If it is described to the shop time be later than it is described go out single time, the incremental time is taken into single duration as described in.
11. it according to the method described in claim 9, it is characterized in that, further includes:
From the History Order data of the trade company of the pending order, extraction History Order details feature, history trade company attribute
Feature and history go out single duration;
History Order details feature, history trade company attributive character and history are gone out into single duration as training sample, using machine
Go out single model described in learning algorithm training.
12. according to the method described in claim 6, it is characterized in that, the multiple action section includes:To user hotline;
The extraction step to the associated dimensional characteristics of user hotline, including:
From the multi-dimensional data, order position feature and the speed effect characteristics of dispatching person are extracted;
The generation step of the action parameter to user hotline, including:
According to the order position attribution, single channel line is sent in planning;
According to the speed effect characteristics and second speed model, one velocity is sent in calculating;
The acquisition step of the execution duration to user hotline, including:
According to it is described send single channel line and it is described send one velocity, get user's duration.
13. according to the method described in claim 6, it is characterized in that, the multiple action section includes:Etc. user hotlines;
The extraction step of the associated dimensional characteristics of user hotlines such as described, including:
From the multi-dimensional data, the History Order data of dispatching person are obtained;
From the History Order data, the historical position of the dispatching person and the band of position of History Order user are extracted;
The generation step of the action parameter of the user hotlines such as described, including:
According to the historical position of the dispatching person and the band of position of the History Order user, count the dispatching person and be located at institute
State the average duration in the band of position of History Order user;
The acquisition step of the execution duration of the user hotlines such as described, including:
According to users' durations such as the average duration, acquisitions.
14. a kind of distribution time estimating device, which is characterized in that including:
Determination unit for determining pending order, has multiple action sections in the processing stream of the pending order;
Unit is estimated, for estimating multiple execution durations corresponding to the multiple action section;
Time processing unit, for according to the multiple execution duration, the estimated delivery time of the acquisition pending order.
15. device according to claim 14, which is characterized in that the time processing unit is specifically used for:
According to the tapeout state of the pending order, determine to estimate initial time;
Initial time is estimated based on described, add up the multiple execution duration, to obtain the estimated delivery time.
16. device according to claim 15, which is characterized in that the time processing unit is additionally operable to:
Initial time and the multiple execution duration are estimated according to described, calculates the multiple action corresponding execution of section
Time started and/or execution end time;
It is suitable according to the priority between the multiple action section corresponding execution time started and/or execution end time
Sequence corrects the multiple execution duration.
17. according to claim 14-16 any one of them devices, which is characterized in that the unit of estimating includes:
Determination subelement, for determining the multiple action, section is corresponding estimates logic;
Subelement is performed, section is corresponding estimates logic for performing the multiple action, to obtain the multiple execution duration.
18. device according to claim 17, which is characterized in that the execution subelement includes:
Extracting sub-module, it is each for from the pending associated multi-dimensional data of order, extracting the multiple action section
The dimensional characteristics of auto correlation;
Submodule is generated, for acting section each associated dimensional characteristics according to the multiple, generates the multiple active region
The respective action parameter of section;
Acquisition submodule for being based on the multiple action respective action parameter of section, obtains the multiple execution duration.
19. device according to claim 18, which is characterized in that the multiple action section includes:To shop section;
The extracting sub-module is specifically used for:From the multi-dimensional data, order position feature and the speed of dispatching person are extracted
Effect characteristics, wherein, the order position feature is the position of each order in the pending affiliated order group of order;
The generation submodule is specifically used for:According to the order position attribution, planning takes single channel line;It is influenced according to the speed
Feature and First Speed model, calculating take one velocity;
The acquisition submodule is specifically used for:According to it is described take single channel line and it is described take one velocity, get shop duration.
20. device according to claim 18, which is characterized in that the multiple action section includes:Take single section;
The extracting sub-module is specifically used for:From the multi-dimensional data, the order detail for extracting the pending order is special
Sign and trade company's attributive character;
The generation submodule is specifically used for:According to the order detail feature and trade company's attributive character, with reference to going out single model,
Generate single duration;
The acquisition submodule is specifically used for:According to lower single time of the pending order and it is described go out single duration, determine
Single time;Estimate the dispatching person of the pending order to the shop time;According to it is described to the shop time and it is described go out single time, obtain
Single duration must be taken.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611069127.4A CN108122042A (en) | 2016-11-28 | 2016-11-28 | Distribution time predictor method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611069127.4A CN108122042A (en) | 2016-11-28 | 2016-11-28 | Distribution time predictor method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108122042A true CN108122042A (en) | 2018-06-05 |
Family
ID=62225472
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611069127.4A Pending CN108122042A (en) | 2016-11-28 | 2016-11-28 | Distribution time predictor method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108122042A (en) |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109146203A (en) * | 2018-09-28 | 2019-01-04 | 北京顺丰同城科技有限公司 | Order distribution information prediction technique, device, electronic equipment and storage medium |
CN109284956A (en) * | 2018-08-10 | 2019-01-29 | 北京三快在线科技有限公司 | A kind of task duration determines method and device and electronic equipment |
CN109461057A (en) * | 2018-11-13 | 2019-03-12 | 拉扎斯网络科技(上海)有限公司 | Order information processing method and order information processing unit |
CN110210946A (en) * | 2019-06-10 | 2019-09-06 | 网易无尾熊(杭州)科技有限公司 | Data processing method and device, medium and calculating equipment |
CN110543968A (en) * | 2019-07-30 | 2019-12-06 | 北京三快在线科技有限公司 | Method and device for determining estimated delivery time, electronic device and storage medium |
CN110706047A (en) * | 2018-07-10 | 2020-01-17 | 北京三快在线科技有限公司 | Method and device for determining delivery information of order |
CN110969381A (en) * | 2018-09-28 | 2020-04-07 | 北京三快在线科技有限公司 | Distribution track segmentation method and device |
CN110969319A (en) * | 2018-09-28 | 2020-04-07 | 北京三快在线科技有限公司 | Assessment duration determination method and device |
CN111325429A (en) * | 2018-12-14 | 2020-06-23 | 中移信息技术有限公司 | Order pushing method, device, medium and equipment |
CN111582574A (en) * | 2020-04-30 | 2020-08-25 | 南京福佑在线电子商务有限公司 | Order unbinding method and device |
CN112085243A (en) * | 2019-06-14 | 2020-12-15 | 达疆网络科技(上海)有限公司 | Calculation method for pre-estimation of delivery time of store order |
CN112101839A (en) * | 2020-07-10 | 2020-12-18 | 上海寻梦信息技术有限公司 | Method for establishing express delivery time prediction model, prediction method and related equipment |
CN112561112A (en) * | 2019-09-10 | 2021-03-26 | 北京三快在线科技有限公司 | Order distribution method and device, computer readable storage medium and electronic equipment |
CN112862133A (en) * | 2019-11-12 | 2021-05-28 | 北京三快在线科技有限公司 | Order processing method and device, readable storage medium and electronic equipment |
CN112990948A (en) * | 2021-04-16 | 2021-06-18 | 北京橙心无限科技发展有限公司 | Service evaluation method, device, equipment, storage medium and computer program product |
CN112990521A (en) * | 2019-12-13 | 2021-06-18 | 顺丰科技有限公司 | Distribution time estimation method and device, computer equipment and storage medium |
CN113033924A (en) * | 2021-05-06 | 2021-06-25 | 杭州师范大学 | Method for predicting delivery duration of package driven by historical data based on Boosting algorithm |
CN113034075A (en) * | 2021-03-29 | 2021-06-25 | 上海寻梦信息技术有限公司 | Logistics waybill timeliness pushing method, system, equipment and storage medium |
CN113222282A (en) * | 2021-05-31 | 2021-08-06 | 北京京东乾石科技有限公司 | Method and device for determining conveying path, readable storage medium and electronic equipment |
CN113780966A (en) * | 2020-06-16 | 2021-12-10 | 北京京东振世信息技术有限公司 | Method and device for generating loading and unloading operation remaining time information |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008087958A (en) * | 2006-09-08 | 2008-04-17 | Canon System Solutions Inc | Information processor, information processing method, program and recording medium |
CN103426075A (en) * | 2013-08-22 | 2013-12-04 | 深圳市华傲数据技术有限公司 | Logistical intelligent pick-up method and system |
CN104504595A (en) * | 2014-12-19 | 2015-04-08 | 上海点啥网络科技有限公司 | Method with function of estimating pick-up time based on online ordering and application thereof |
CN105719110A (en) * | 2015-05-22 | 2016-06-29 | 北京小度信息科技有限公司 | Order processing method and device |
-
2016
- 2016-11-28 CN CN201611069127.4A patent/CN108122042A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008087958A (en) * | 2006-09-08 | 2008-04-17 | Canon System Solutions Inc | Information processor, information processing method, program and recording medium |
CN103426075A (en) * | 2013-08-22 | 2013-12-04 | 深圳市华傲数据技术有限公司 | Logistical intelligent pick-up method and system |
CN104504595A (en) * | 2014-12-19 | 2015-04-08 | 上海点啥网络科技有限公司 | Method with function of estimating pick-up time based on online ordering and application thereof |
CN105719110A (en) * | 2015-05-22 | 2016-06-29 | 北京小度信息科技有限公司 | Order processing method and device |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110706047A (en) * | 2018-07-10 | 2020-01-17 | 北京三快在线科技有限公司 | Method and device for determining delivery information of order |
CN110706047B (en) * | 2018-07-10 | 2021-12-28 | 北京三快在线科技有限公司 | Method and device for determining delivery information of order |
CN109284956A (en) * | 2018-08-10 | 2019-01-29 | 北京三快在线科技有限公司 | A kind of task duration determines method and device and electronic equipment |
CN110969381A (en) * | 2018-09-28 | 2020-04-07 | 北京三快在线科技有限公司 | Distribution track segmentation method and device |
CN110969319A (en) * | 2018-09-28 | 2020-04-07 | 北京三快在线科技有限公司 | Assessment duration determination method and device |
CN110969381B (en) * | 2018-09-28 | 2022-03-08 | 北京三快在线科技有限公司 | Distribution track segmentation method and device |
CN109146203A (en) * | 2018-09-28 | 2019-01-04 | 北京顺丰同城科技有限公司 | Order distribution information prediction technique, device, electronic equipment and storage medium |
CN109461057A (en) * | 2018-11-13 | 2019-03-12 | 拉扎斯网络科技(上海)有限公司 | Order information processing method and order information processing unit |
CN111325429A (en) * | 2018-12-14 | 2020-06-23 | 中移信息技术有限公司 | Order pushing method, device, medium and equipment |
CN110210946A (en) * | 2019-06-10 | 2019-09-06 | 网易无尾熊(杭州)科技有限公司 | Data processing method and device, medium and calculating equipment |
CN112085243A (en) * | 2019-06-14 | 2020-12-15 | 达疆网络科技(上海)有限公司 | Calculation method for pre-estimation of delivery time of store order |
CN110543968A (en) * | 2019-07-30 | 2019-12-06 | 北京三快在线科技有限公司 | Method and device for determining estimated delivery time, electronic device and storage medium |
CN110543968B (en) * | 2019-07-30 | 2022-04-12 | 北京三快在线科技有限公司 | Method and device for determining estimated delivery time, electronic device and storage medium |
WO2021017609A1 (en) * | 2019-07-30 | 2021-02-04 | 北京三快在线科技有限公司 | Determination of estimated time of arrival |
CN112561112A (en) * | 2019-09-10 | 2021-03-26 | 北京三快在线科技有限公司 | Order distribution method and device, computer readable storage medium and electronic equipment |
CN112862133A (en) * | 2019-11-12 | 2021-05-28 | 北京三快在线科技有限公司 | Order processing method and device, readable storage medium and electronic equipment |
CN112990521A (en) * | 2019-12-13 | 2021-06-18 | 顺丰科技有限公司 | Distribution time estimation method and device, computer equipment and storage medium |
CN111582574B (en) * | 2020-04-30 | 2020-12-15 | 南京福佑在线电子商务有限公司 | Order unbinding method and device |
CN111582574A (en) * | 2020-04-30 | 2020-08-25 | 南京福佑在线电子商务有限公司 | Order unbinding method and device |
CN113780966B (en) * | 2020-06-16 | 2023-09-26 | 北京京东振世信息技术有限公司 | Method and device for generating loading and unloading operation residual time information |
CN113780966A (en) * | 2020-06-16 | 2021-12-10 | 北京京东振世信息技术有限公司 | Method and device for generating loading and unloading operation remaining time information |
CN112101839A (en) * | 2020-07-10 | 2020-12-18 | 上海寻梦信息技术有限公司 | Method for establishing express delivery time prediction model, prediction method and related equipment |
CN113034075A (en) * | 2021-03-29 | 2021-06-25 | 上海寻梦信息技术有限公司 | Logistics waybill timeliness pushing method, system, equipment and storage medium |
CN112990948A (en) * | 2021-04-16 | 2021-06-18 | 北京橙心无限科技发展有限公司 | Service evaluation method, device, equipment, storage medium and computer program product |
CN113033924A (en) * | 2021-05-06 | 2021-06-25 | 杭州师范大学 | Method for predicting delivery duration of package driven by historical data based on Boosting algorithm |
CN113222282A (en) * | 2021-05-31 | 2021-08-06 | 北京京东乾石科技有限公司 | Method and device for determining conveying path, readable storage medium and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108122042A (en) | Distribution time predictor method and device | |
CN106549772B (en) | Resource prediction method, system and capacity management device | |
US20150058078A1 (en) | Rule to constraint translator for business application systems | |
US9058216B2 (en) | Method for estimating job run time | |
US20200057918A1 (en) | Systems and methods for training artificial intelligence to predict utilization of resources | |
US8374912B2 (en) | System and method for managing and optimizing advertising campaigns managed on the internet | |
US11301794B2 (en) | Machine for labor optimization for efficient shipping | |
US11567807B2 (en) | Allocation of shared computing resources using source code feature extraction and machine learning | |
Tiemessen et al. | Dynamic demand fulfillment in spare parts networks with multiple customer classes | |
US11281986B2 (en) | Estimating utilization of network resources using time series data | |
Lu et al. | Show me the money: Dynamic recommendations for revenue maximization | |
CN109711887A (en) | Generation method, device, electronic equipment and the computer media of store recommendation list | |
WO2018232331A1 (en) | Systems and methods for optimizing and simulating webpage ranking and traffic | |
CN109685589A (en) | Method and apparatus for pushed information | |
CN109697637A (en) | Object type determines method, apparatus, electronic equipment and computer storage medium | |
US20220147669A1 (en) | Scalable Modeling for Large Collections of Time Series | |
CN108960537A (en) | The prediction technique and device of logistics order, readable medium | |
CN108777701A (en) | A kind of method and device of determining receiver | |
CN109741177A (en) | Appraisal procedure, device and the intelligent terminal of user credit | |
CN109636227A (en) | Method for allocating tasks, device, electronic equipment and computer readable storage medium | |
CN109583799A (en) | The method and device of region division, electronic equipment | |
Kolková et al. | Demand forecasting: an alternative approach based on technical indicator Pbands | |
CN111798280A (en) | Multimedia information recommendation method, device and equipment and storage medium | |
Wang et al. | The mobile production vehicle routing problem: Using 3D printing in last mile distribution | |
CN105654326A (en) | Information processing system and information processing method |
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: 20180605 |
|
RJ01 | Rejection of invention patent application after publication |