CN109685275A - Distribution team load pressure prediction method and device, electronic equipment and storage medium - Google Patents
Distribution team load pressure prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109685275A CN109685275A CN201811607316.1A CN201811607316A CN109685275A CN 109685275 A CN109685275 A CN 109685275A CN 201811607316 A CN201811607316 A CN 201811607316A CN 109685275 A CN109685275 A CN 109685275A
- Authority
- CN
- China
- Prior art keywords
- load pressure
- team
- prediction
- prediction model
- dispatching
- 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 49
- 238000012549 training Methods 0.000 claims abstract description 132
- 238000012360 testing method Methods 0.000 claims description 199
- 230000015654 memory Effects 0.000 claims description 25
- 238000013215 result calculation Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims 1
- 230000010365 information processing Effects 0.000 abstract description 2
- 238000012512 characterization method Methods 0.000 description 11
- 230000002159 abnormal effect Effects 0.000 description 7
- 230000005856 abnormality Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 235000013399 edible fruits Nutrition 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005055 memory storage Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 235000007926 Craterellus fallax Nutrition 0.000 description 1
- 240000007175 Datura inoxia Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012956 testing procedure Methods 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention relates to the technical field of information processing, and discloses a distribution team load pressure prediction method and device, electronic equipment and a storage medium. In the invention, second load pressure characteristic data of a delivery team for prediction is obtained; and predicting the load pressure of the distribution team at the preset time according to the second load pressure characteristic data and a prediction model, wherein the prediction model is obtained by training according to the collected first load pressure characteristic data of the distribution team in advance. Because the predicted load pressure is the load pressure of a distribution team, the comprehensive load pressure of a certain distribution area or a certain distribution station to which the team belongs can be evaluated, so that the distribution quality is ensured, and the distribution efficiency is improved. And the prediction model can make detailed prediction adjustment aiming at different teams, so that the prediction accuracy is improved.
Description
Technical field
The present invention relates to technical field of information processing, in particular to a kind of dispatching team load pressure prediction technique.
Background technique
Currently, dispatching platform needs to calculate each dispatching in real time in dispatching business for scheduling strategy of making rational planning for
The load pressure of personnel.When dispatching platform monitors the load pressure appearance exception of each dispatching personnel, can make a gift to someone according to respectively matching
The current load pressure of member, which makes next transport power and order volume etc., to be reasonably adjusted, from the load for making each dispatching personnel
Pressure is more balanced.
Inventors have found that in the case that single amount in certain region increases or reduces suddenly, also according to real-time load pressure
Power is scheduled next transport power, and the dispatching quality in next a period of time is caused not can guarantee.
Summary of the invention
Embodiment of the present invention is designed to provide a kind of dispatching team load pressure prediction technique, can to adjust
Spending platform, to improve dispatching efficiency, can guarantee dispatching quality according to prediction result Optimized Operation strategy.
In order to solve the above technical problems, embodiments of the present invention provide a kind of load pressure prediction side, dispatching team
Method, comprising: obtain the second load pressure characteristic of the dispatching team for prediction;According to the second load pressure characteristic
According to and prediction model, to dispatching, team is predicted in the load pressure of preset time, wherein prediction model be previously according to
What the first load pressure characteristic training of the dispatching team of acquisition obtained.
Embodiments of the present invention additionally provide a kind of dispatching team load pressure prediction meanss, comprising: module is obtained,
For obtaining the second load pressure characteristic for being used for the dispatching team of prediction;Prediction module, for being pressed according to the second load
Power characteristic and prediction model, to dispatching, team is predicted in the load pressure of preset time, wherein prediction model is
What the first load pressure characteristic training previously according to the dispatching team of acquisition obtained.
Embodiments of the present invention additionally provide a kind of electronic equipment, including memory and processor, memory storage meter
Calculation machine program, processor execute the second load pressure characteristic for obtaining the dispatching team for prediction when running program;Root
According to the second load pressure characteristic and prediction model, dispatching team is predicted in the load pressure of preset time,
In, prediction model be previously according to the dispatching team of acquisition the first load pressure characteristic training obtain.
Embodiments of the present invention additionally provide a kind of non-volatile memory medium, for storing computer-readable program,
Computer-readable program is for executing dispatching team load pressure prediction technique as above for computer.
In terms of existing technologies, the main distinction and its effect are embodiment of the present invention: by negative according to second
Pressure characterization data and prediction model are carried, dispatching team is predicted in the load pressure of preset time, may make that scheduling is flat
Platform obtains each team of preset time in time and obtains load pressure situation, and is ready work in advance according to the load pressure predicted
Make, adjust scheduling strategy, so that order is handled in time, due to the load that the load pressure of prediction is a dispatching team
Pressure, the integrated load pressure that region or some dispatching website can be dispensed to some belonging to team is assessed, to protect
Dispatching quality has been demonstrate,proved, dispatching efficiency is improved.And prediction model can make careful prediction adjustment for different team,
Improve prediction accuracy.Further, since prediction model is the first load pressure characteristic previously according to the dispatching team of acquisition
It is obtained according to training, i.e., for training data from the dispatching true historical data of team, reference value is high, and prediction may make to tie
Fruit is relatively reliable.
In addition, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, specifically: with predetermined period, pre-
If the first load pressure characteristic of acquisition dispatching team, obtained L item the first load pressure characteristic, L are in duration
The amount of cycles for including in preset duration;First load pressure characteristic included at least team's number, dispatching team in the past
The average load pressure of a period of time;Training prediction model, specifically: using L item the first load pressure characteristic as training
Collection, training prediction model.
In addition, the first load pressure characteristic further includes following any combination: dispatching team where urban information,
Dispense team's attribute information, timeslice information, weather pattern, temperature.So that the data referred to during training pattern are more
Abundant, the prediction result of model is more reliable.
In addition, above-mentioned second load pressure characteristic includes the timeslice to be predicted for specifying above-mentioned preset time.
Method by the way that timeslice to be predicted is arranged, divides timeslice for future time, following according to timeslice prediction to be predicted
Load pressure can predict the load pressure of some the following time and need not train needle during training pattern
To the prediction model of the load pressure data of the following different moments, so that scheme has higher feasibility.
In addition, the first load pressure characteristic of the dispatching team previously according to acquisition, training prediction model, comprising:
Acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, obtains the dispatching team for corresponding to each duration
The first load pressure characteristic;N is the natural number greater than 1;The first load pressure according to the dispatching team of each duration is special
Data are levied, prediction model are trained respectively, the prediction model after obtaining N number of training;To the prediction model after N number of training
It is tested;The prediction model after the smallest training of error will be predicted, as target prediction model;According to the second load pressure
Characteristic and prediction model, to dispatching, team is predicted in the load pressure of preset time, specifically: it is negative according to second
It carries pressure characterization data and target prediction model and dispatching team is predicted in the load pressure of preset time.Training is multiple
Prediction model, and using the smallest prediction model of prediction error obtained after training as target training pattern, it is instructed using target
Practice model to predict load pressure, may make the error for predicting the load pressure and actual loading pressure come as small as possible, in advance
It is more accurate to survey result.
In addition, the above-mentioned prediction model to after N number of training is tested, specifically include: obtaining N number of test set;N number of survey
Examination collection is corresponded with the prediction model after N number of training;According to N number of test set respectively to the prediction model after corresponding training
It is tested;According to the prediction error of the prediction model after test result calculations training.Obtain N number of test set, respectively to this
N number of test set corresponds training set and is tested, it can be achieved that the personalization to prediction model is tested, according to test result meter
The test error of each prediction model is calculated, to obtain the more accurate prediction model of prediction result.
In addition, above-mentioned test result obtains in the following manner: when testing the test data in test set,
The prediction result of prediction model output after training is compared with the legitimate reading of test data, if error is pre- first
If in range, then the test result of discriminating test data is correct, the otherwise test result mistake of discriminating test data.Setting is pre-
Survey result error range, when error within a preset range when, determine that the test result of this test data is correct, due to mould
The prediction result of type be difficult it is completely the same with legitimate reading, as long as the prediction result of model and the error of legitimate reading can connect
By in the range of, then it is assumed that the prediction result of model is correct, so that prediction model has certain fault-tolerance.
In addition, above-mentioned test result obtains in the following manner: when testing the test data in test set,
It is 0 or 1 according to the test result specification that preset rules export the prediction model after training;Number will be tested according to preset rules
Legitimate reading specification in is 0 or 1;If the test result after specification is identical as legitimate reading, the survey of discriminating test data
Test result is correct, otherwise the test prediction result mistake of discriminating test data.Since the rule of specification is identical, if after specification
Test result is identical as legitimate reading, can discriminating test result it is correct so that test result compared with legitimate reading more
Simply, predict that the calculating of error is also more convenient.
In addition, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, is the dispatching after rejecting abnormalities data
First load pressure characteristic of team: where abnormal data is the load pressure spy for meeting the dispatching team of preset condition
Levy data.Rejecting abnormalities data can avoid influence of the abnormal data to the model structure trained, so that the prediction result of model
It is more accurate.
Detailed description of the invention
Fig. 1 is the dispatching team load pressure prediction technique flow chart in first embodiment according to the present invention;
Fig. 2 is the dispatching team load pressure prediction technique flow chart in second embodiment according to the present invention;
Fig. 3 is the testing procedure flow chart in third embodiment according to the present invention;
Fig. 4 is the dispatching team load pressure prediction meanss schematic diagram in the 4th embodiment according to the present invention;
Fig. 5 is the electronic devices structure schematic diagram that the 5th embodiment provides according to the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Each embodiment be explained in detail.However, it will be understood by those skilled in the art that in each implementation of the invention
In mode, in order to make the reader understand this application better, many technical details are proposed.But even if without these technologies
The application technical side claimed also may be implemented in details and various changes and modifications based on the following respective embodiments
Case.The division of each embodiment is for convenience, should not to constitute any limit to specific implementation of the invention below
Fixed, each embodiment can be combined with each other mutual reference under the premise of reconcilable.
The first embodiment of the present invention is related to a kind of dispatching team load pressure prediction techniques.Team's load pressure can
To be interpreted as total single summation measured divided by the single ability of jockey's maximum each in team back of team, if the load pressure calculated
Value is greater than 1, then it represents that excess load back is single for the team, unbearable more single amounts.In the present embodiment, due to prediction
Load pressure is the load pressure of a dispatching team, so as to some dispatching region belonging to team or some dispatching
The integrated load pressure of website is assessed, to ensure that dispatching quality, improves dispatching efficiency.And model can be directed to
Different team makes careful prediction adjustment, improves prediction accuracy.As shown in Figure 1, this method comprises:
Step S101 obtains the second load pressure characteristic of the dispatching team for prediction.
Specifically, the dispatching team for prediction can be obtained from dispatching platform or human-computer interaction interface as needed
Second load pressure characteristic, wherein the second load pressure characteristic include for specify preset time it is to be predicted when
Between piece.Timeslice can be understood as a unit time, such as be divided into a timeslice for 15 minutes.Timeslice to be predicted
It can be used to some following preset time of instruction, for example, needing to predict 11:15 points of load if current time was 11 o'clock sharps
If desired pressure predicts 11:30 points of load pressure then 1 can be set by timeslice to be predicted, then can will be to
Predicted time piece is set as 2, and so on.
In the present embodiment, is specified, can be predicted not by the way that the time that timeslice to be predicted predicts needs is arranged
Carry out the load pressure of some time, and it is not necessary that prediction model is respectively trained for following different moments so that scheme have compared with
High feasibility.
Step S102 bears dispatching team in preset time according to the second load pressure characteristic and prediction model
Pressure is carried to be predicted.
In one example, current point in time 11:00, the second load pressure characteristic include team's number 355,
The single amount 300 of maximum back that team can bear, weather pattern is encoded to 1 (rainfall), and the dispatching personnel amount of team's current active is
20, the timeslice of prediction is 1 (every 15 minutes are a timeslice), according to these the second load pressure characteristics and in advance
The prediction model of acquisition, can be predicted and number the team for being 355 out in the load pressure of 11:15 is M, and dispatching platform can be according to this
A load pressure value M is adjusted scheduling strategy.
In the present embodiment, prediction model is the first load pressure characteristic previously according to the dispatching team of acquisition
It obtains according to training, below to the prediction model of present embodiment, is specifically described.
XGB model (i.e. xgboost model) can be used in the prediction model of present embodiment, and XGB model can increase prediction
The robustness of model, compared to traditional machine learning algorithm, speed is fast, effect is good, can handle large-scale data and support
Multilingual and customized loss function.
For training the first load pressure characteristic of the dispatching team of prediction model, can adopt in the following manner
Collection obtains: with predetermined period, in preset duration acquisition dispatching team the first load pressure characteristic, L articles obtained the
One load pressure characteristic, L be preset duration in include amount of cycles, using L item the first load pressure characteristic as
Training set, training prediction model.For example, the first load of each dispatching team was acquired for the period with 15 minutes within a week
7x24x (60/15)=672 the first load pressure characteristic of each dispatching team then can be obtained in pressure characterization data, with
672 load pressure characteristics of each team are as training set, training prediction model.
Wherein, the first load pressure characteristic includes at least team's number, dispatching team puts down the past period
Equal load pressure may also include urban information, dispatching team's attribute information, timeslice information, weather class where dispensing team
Any combination in the dimensions such as type, temperature, wherein dispatching team's attribute information may include dispatching team's registration time length, dispatching group
The single amount of maximum back etc. that team can bear.
In the concrete realization, the timeslice information in the first load pressure characteristic can be established in the following manner:
By taking the unit time is 15 minutes as an example, can be divided into several timeslices one day time, 00:00 to 00:15 is first
A timeslice;00:15 to 00:30 is second timeslice, and so on.That is, this moment can obtain in 00:15
First timeslice, the number of first timeslice are 1, can obtain second timeslice at this moment of 00:30, second
The number of timeslice is 2, and so on.
It is appreciated that the urban information where dispatching team, can effectively reflect geographical position locating for dispatching team
It sets;Dispatching team's attribute information can effectively reflect team's characteristic of dispatching team, enable model for different
Team makes careful prediction adjustment, and the combination of the characteristic by these different dimensions, may make training pattern
The data referred in the process more horn of plenty, the prediction result of model are more reliable.
In practical applications, the characteristic for each dimension for including in the first load pressure characteristic can be such as table 1
It is shown:
Table 1
In the concrete realization, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, can also be that rejecting is different
First load pressure characteristic of the dispatching team after regular data: where abnormal data is the dispatching group for meeting preset condition
The load pressure characteristic of team, for example, preset condition is that average daily single measure is greater than 300, then average daily single amount is less than or equal to 300
The data of Small Groups can be removed, largely less than normal to avoid the load pressure due to Small Groups, and prediction model is caused to be distorted
Problem.
It is appreciated that the output of prediction model, the load pressure as predicted, the input of prediction model are x1To xn, indicate
The characteristic of different dimensions, during carrying out prediction model training, by constantly comparing the input according to prediction model
The output valve being calculated, true value corresponding to the input with prediction model are (true corresponding to the characteristic i.e. with input
Real load pressure), the parameter of prediction model is adjusted, thus the prediction model after being trained.
In terms of existing technologies, the main distinction and its effect are embodiment of the present invention: by negative according to second
Pressure characterization data and prediction model are carried, dispatching team is predicted in the load pressure of preset time, may make that scheduling is flat
Platform obtains each team of preset time in time and obtains load pressure situation, and is ready work in advance according to the load pressure predicted
Make, adjust scheduling strategy, so that order is handled in time, due to the load that the load pressure of prediction is a dispatching team
Pressure, the integrated load pressure that region or some dispatching website can be dispensed to some belonging to team is assessed, to protect
Dispatching quality has been demonstrate,proved, dispatching efficiency is improved.And prediction model can make careful prediction adjustment for different team,
Improve prediction accuracy.Further, since prediction model is the first load pressure feature previously according to the dispatching team of acquisition
Data training obtains, i.e., for training data from the dispatching true historical data of team, reference value is high, may make prediction
As a result relatively reliable.
Second embodiment of the present invention is related to a kind of dispatching team load pressure prediction technique, in second embodiment
A kind of acquisition modes of specific prediction model are provided, the present embodiment flow chart is as shown in Fig. 2, specifically described below:
Step S100, obtains prediction model in advance.The acquisition modes of prediction model in present embodiment, especially by
Step S1011 to step S1014 is realized, specifically described below.
Step S1011, acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, is corresponded to
First load pressure characteristic of the dispatching team of each duration.
Specifically, above-mentioned N is the natural number greater than 1.By taking a week is a duration as an example, to 1 day to 11 November
Months 7 days, November 15 to November 21, the first of acquisition dispatching team respectively in these three weeks on November 29 to December 5
Load pressure characteristic, wherein in the data dimension and first embodiment that the first load pressure characteristic includes substantially
It is identical, to avoid repeating, no longer repeat one by one here.
Step S1012, according to the first load pressure characteristic of the dispatching team of each duration, respectively to prediction model
It is trained, the prediction model after obtaining N number of training.
Still negative according to the first of the dispatching team in each week in these three weeks by taking above three week as an example
Pressure characterization data is carried, prediction model is trained respectively, the prediction model after obtaining three training, i.e., extremely by November 1
The first prediction model that the first load pressure characteristic on November 7 trains, by November 15 to November 21 first
The second prediction model that load pressure characteristic trains, by the first load pressure characteristic in November 29 to December 5
According to the third prediction model trained.
Step S1013 tests the prediction model after N number of training.That is, being obtained after being trained to three
Prediction model tested respectively.
Step S1014 will predict the prediction model after the smallest training of error, as target prediction model.
It, can be according to prediction result and legitimate reading specifically, after testing respectively three prediction models
Gap calculates the prediction error of each prediction model, and will predict the smallest prediction model of error, as target training mould
This target prediction model is saved as file by type, when needing to predict load pressure, calls this document, with this document storage
Target prediction model predicts load pressure.
For example, the prediction error of the first prediction model is x, the prediction error of the second prediction model is y, and third predicts mould
The prediction error of type is z, wherein x > y > z, i.e. the error of third prediction model is minimum, then pre- as target using third prediction model
Model is surveyed, and this target prediction model is saved as into file.
In the present embodiment, after obtaining target prediction model, load pressure is predicted according to target prediction model
Process is roughly the same with first embodiment, to avoid repeating, here no longer to the implementation detail of step S101 and step S102
It repeats one by one.
Step S101 obtains the second load pressure characteristic of the dispatching team for prediction.
Step S102 bears dispatching team in preset time according to the second load pressure characteristic and prediction model
Pressure is carried to be predicted.
Present embodiment compared with the prior art for, multiple prediction models of training, and the prediction error that will be obtained after training
The smallest prediction model predicts load pressure using target training pattern as target training pattern, may make to predict and
Load pressure and actual loading pressure error it is as small as possible, prediction result is more accurate.
Third embodiment of the present invention is related to a kind of dispatching team load pressure prediction technique, present embodiment and the
Two embodiments are roughly the same, and the main distinction is: in the present embodiment, provide it is a kind of specifically to prediction model into
The method of row test, the step in present embodiment is roughly the same with the step in second embodiment, and to step S1031
Further refinement has been done, as shown in figure 3, the refinement step of step 1031 is specifically described below:
Step S10131 obtains N number of test set.
Specifically, the prediction model after N number of test set and N number of training corresponds, for example, with a week
The first load pressure characteristic as a training set, a prediction model can be trained, acquired for three weeks respectively
First load pressure characteristic can train three prediction models, obtain survey corresponding with these three prediction models respectively
Examination collection.
As shown in table 2, test set corresponding with the training set in November 1 to November 7 is November 8 to November 14
Collected load pressure characteristic in this week, the feature dimensions that the load pressure characteristic in the test set is included
It spends identical as the characteristic dimension that the first load pressure characteristic in the training set in November 1 to November 7 is included.It adopts
The mode of collection, can also be identical as the acquisition mode of the first load pressure characteristic in training set, i.e., with predetermined period,
The load pressure characteristic of acquisition dispatching team, obtained a plurality of load pressure are special in this week on November 8 to November 14
Data are levied as test data, the prediction model for obtaining to the training set training by November 1 to November 7 is surveyed
Examination.Similarly, test set corresponding with the training set in November 15 to November 21 be November 22 to November 28 this
Collected load pressure characteristic in week;Test set corresponding with the training set in November 29 to December 5 is December 6
The collected load pressure characteristic day on December 12 in this week.Each load pressure characteristic in test set
The characteristic dimension for being included, the characteristic dimension for being included with each item the first load pressure characteristic in corresponding training set
It is identical, each load pressure characteristic in test set be used for the prediction model obtained by the training of corresponding training set into
Row test.
Table 2
Training set | Test set |
11-01~11-07 | 11-08~11-14 |
11-15~11-21 | 11-22~11-28 |
11-29~12-05 | 12-06~12-12 |
Step S10132 respectively tests the prediction model after corresponding training according to N number of test set.
Specifically, the load pressure characteristic in test set data is made when testing each prediction model
For the input of prediction model, the load pressure of prediction is calculated using prediction model.
Step S10133, according to the prediction error of the prediction model after test result calculations training.
Specifically, after the load pressure for calculating prediction using prediction model, can by the load pressure value of prediction with
Whether true load pressure value is compared, to judge prediction result correctly to get to test result.It is every calculating
When the prediction error of a prediction model, the test data item number of all test result mistakes is counted, the prediction model can be obtained
Prediction error.
In a specific example, test result obtains in the following manner: to the test data in test set into
When row test, the prediction result of the prediction model output after training is compared with the legitimate reading of test data, if error
In the first preset range (such as float up and down 10%), then the test result of discriminating test data is correct, otherwise discriminating test
The test result mistake of data.Due to the prediction result of model be difficult it is completely the same with legitimate reading, by setting prediction
As a result error range, when model prediction result and legitimate reading error within the acceptable range, then it is assumed that model
Prediction result is correct, so that prediction model has certain fault-tolerance.
In another specific example, test result obtains in the following manner: to the test data in test set
It is 0 or 1 according to the test result specification that preset rules export the prediction model after training when being tested;According to default rule
It is then 0 or 1 by the legitimate reading specification in test data;If the test result after specification is identical as legitimate reading, determine to survey
The test result for trying data is correct, otherwise the test result mistake of discriminating test data.For example, the rule of specification are as follows: be greater than
0.5 specification is 1, and otherwise specification is 0, if prediction result is 0.9, legitimate reading 0.7, then and prediction result and legitimate reading
All it is 1 by specification, can determine that test result is correct.Since the rule of specification is identical, if test result and true knot after specification
Fruit is identical, can discriminating test result it is correct so that test result is simpler compared with legitimate reading, predict error
It calculates also more convenient.
Present embodiment compared with the prior art for, different test sets is respectively adopted to each prediction model and is surveyed
The personalization test, it can be achieved that prediction model is tried, according to the test error of each prediction model of test result calculations, so that
The prediction effect quality of each prediction model has a unified module.
The step of various methods divide above, be intended merely to describe it is clear, when realization can be merged into a step or
Certain steps are split, multiple steps are decomposed into, as long as including identical logical relation, all in the protection model of this patent
In enclosing;To adding inessential modification in algorithm or in process or introducing inessential design, but its calculation is not changed
The core design of method and process is all in the protection scope of the patent.
4th embodiment of the invention is related to a kind of dispatching team load pressure prediction meanss, as shown in figure 4, the dress
It sets and includes:
Module 401 is obtained, for obtaining the second load pressure characteristic for being used for the dispatching team of prediction;
Prediction module 402, for being preset to dispatching team according to the second load pressure characteristic and prediction model
The load pressure of time predicted, wherein, prediction model be previously according to the dispatching team of acquisition the first load pressure it is special
Sign data training obtains.
In one example, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, specially with default week
Phase, acquisition dispenses the first load pressure characteristic of team, obtained L item the first load pressure feature in preset duration
Data, L are the amount of cycles for including in preset duration;First load pressure characteristic includes at least team's number, dispatching group
Average load pressure of the team in the past period;Training prediction model, specifically: with L item the first load pressure characteristic
As training set, training prediction model.
In one example, the characteristic of above-mentioned at least two dimension includes following any combination: where dispatching team
Urban information, dispatching team's attribute information, timeslice information, weather pattern, temperature.
In one example, above-mentioned second load pressure characteristic include for specify preset time it is to be predicted when
Between piece.
In one example, previously according to the first load pressure characteristic of the dispatching team of acquisition, training prediction mould
Type, comprising: acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, obtains corresponding to each duration
Dispense the first load pressure characteristic of team;N is the natural number greater than 1;It is negative according to the first of the dispatching team of each duration
Pressure characterization data is carried, prediction model is trained respectively, the prediction model after obtaining N number of training;After N number of training
Prediction model is tested;The prediction model after the smallest training of error will be predicted, as target prediction model;According to second
Load pressure characteristic and prediction model, to dispatching, team is predicted in the load pressure of preset time, specifically: root
Dispatching team is predicted in the load pressure of preset time according to the second load pressure characteristic and target prediction model.
In one example, the above-mentioned prediction model to after N number of training is tested, and is specifically included: obtaining N test
Collection;Prediction model after N number of test set and N number of training corresponds;According to N number of test set respectively to corresponding training after
Prediction model is tested;According to the prediction error of the prediction model after test result calculations training.
In one example, above-mentioned test result obtains in the following manner: carrying out to the test data in test set
When test, the prediction result of the prediction model output after training is compared with the legitimate reading of test data, if error exists
In first preset range, then the test result of discriminating test data is correct, otherwise the test result mistake of discriminating test data.
In one example, above-mentioned test result obtains in the following manner: carrying out to the test data in test set
It is 0 or 1 according to the test result specification that preset rules export the prediction model after training when test;It will according to preset rules
Legitimate reading specification in test data is 0 or 1;If the test result after specification is identical as legitimate reading, discriminating test number
According to test result it is correct, the otherwise test prediction result mistake of discriminating test data.
In one example, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, is rejecting abnormalities data
First load pressure characteristic of dispatching team afterwards: where abnormal data be meet preset condition dispatching team it is negative
Carry pressure characterization data.
In one example, the type of above-mentioned prediction model is XGB model.
It is not difficult to find that present embodiment be with first embodiment to the corresponding Installation practice of third embodiment,
Present embodiment can work in coordination implementation with first embodiment to third embodiment.First embodiment to third is implemented
The relevant technical details mentioned in mode are still effective in the present embodiment, and in order to reduce repetition, which is not described herein again.Phase
Ying Di, the relevant technical details mentioned in present embodiment are also applicable in first embodiment into third embodiment.
It is noted that each module involved in present embodiment is logic module, in practical applications,
One logic unit can be a physical unit, be also possible to a part of a physical unit, can also be with multiple physics
The combination of unit is realized.In addition, in order to protrude innovative part of the invention, it will not be with this hair of solution in present embodiment
The technical issues of bright proposed, the less close unit of relationship introduced, but this does not indicate that there is no other in present embodiment
Unit.
5th embodiment of the invention is related to a kind of electronic equipment, as shown in figure 5, the electronic equipment includes: at least one
A processor 501;And the memory 502 with the communication connection of at least one processor 501;And it is communicated with scanning means
The communication component 503 of connection, communication component 503 send and receive data under the control of processor 501;Wherein, memory
502 are stored with the instruction that can be executed by least one processor 501, and instruction is executed by least one processor 501 to realize:
Obtain the second load pressure characteristic of the dispatching team for prediction;According to the second load pressure characteristic
According to and prediction model, to dispatching, team is predicted in the load pressure of preset time, wherein prediction model be previously according to
What the first load pressure characteristic training of the dispatching team of acquisition obtained.
In one example, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, specially with default week
Phase, acquisition dispenses the first load pressure characteristic of team, obtained L item the first load pressure feature in preset duration
Data, L are the amount of cycles for including in preset duration;First load pressure characteristic includes at least team's number, dispatching group
Average load pressure of the team in the past period;Training prediction model, specifically: with L item the first load pressure characteristic
As training set, training prediction model.
In one example, the characteristic of above-mentioned at least two dimension includes following any combination: where dispatching team
Urban information, dispatching team's attribute information, timeslice information, weather pattern, temperature.
In one example, above-mentioned second load pressure characteristic include for specify preset time it is to be predicted when
Between piece.
In one example, previously according to the first load pressure characteristic of the dispatching team of acquisition, training prediction mould
Type, comprising: acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, obtains corresponding to each duration
Dispense the first load pressure characteristic of team;N is the natural number greater than 1;It is negative according to the first of the dispatching team of each duration
Pressure characterization data is carried, prediction model is trained respectively, the prediction model after obtaining N number of training;After N number of training
Prediction model is tested;The prediction model after the smallest training of error will be predicted, as target prediction model;According to second
Load pressure characteristic and prediction model, to dispatching, team is predicted in the load pressure of preset time, specifically: root
Dispatching team is predicted in the load pressure of preset time according to the second load pressure characteristic and target prediction model.
In one example, the above-mentioned prediction model to after N number of training is tested, and is specifically included: obtaining N number of test
Collection;Prediction model after N number of test set and N number of training corresponds;According to N number of test set respectively to corresponding training after
Prediction model is tested;According to the prediction error of the prediction model after test result calculations training.
In one example, above-mentioned test result obtains in the following manner: carrying out to the test data in test set
When test, the prediction result of the prediction model output after training is compared with the legitimate reading of test data, if error exists
In first preset range, then the test result of discriminating test data is correct, otherwise the test result mistake of discriminating test data.
In one example, above-mentioned test result obtains in the following manner: carrying out to the test data in test set
It is 0 or 1 according to the test result specification that preset rules export the prediction model after training when test;It will according to preset rules
Legitimate reading specification in test data is 0 or 1;If the test result after specification is identical as legitimate reading, discriminating test number
According to test result it is correct, the otherwise test prediction result mistake of discriminating test data.
In one example, the first load pressure characteristic of the dispatching team of above-mentioned acquisition, is rejecting abnormalities data
First load pressure characteristic of dispatching team afterwards: where abnormal data be meet preset condition dispatching team it is negative
Carry pressure characterization data.
In one example, the type of above-mentioned prediction model is XGB model.
Specifically, which includes: one or more processors 501 and memory 502, at one in Fig. 5
For reason device 501.Processor 501, memory 502 can be connected by bus or other modes, to pass through bus in Fig. 5
For connection.Memory 502 is used as a kind of non-volatile computer readable storage medium storing program for executing, can be used for storing non-volatile software
Program, non-volatile computer executable program and module.Processor 501 is stored in non-in memory 502 by operation
Volatibility software program, instruction and module are realized thereby executing the various function application and data processing of equipment
State dispatching team's load pressure prediction technique.
Memory 502 may include storing program area and storage data area, wherein storing program area can store operation system
Application program required for system, at least one function;It storage data area can the Save option list etc..In addition, memory 502 can
To include high-speed random access memory, can also include nonvolatile memory, a for example, at least disk memory,
Flush memory device or other non-volatile solid state memory parts.In some embodiments, memory 502 it is optional include relative to
The remotely located memory 502 of processor 501, these remote memories 502 can pass through network connection to external equipment.On
The example for stating network includes but is not limited to internet, intranet, local area network, mobile radio communication and combinations thereof.
One or more module is stored in memory 502, when being executed by one or more processor 501, is held
Dispatching team load pressure prediction technique in the above-mentioned any means embodiment of row.
Method provided by the application embodiment can be performed in the said goods, has the corresponding functional module of execution method
And beneficial effect, the not technical detail of detailed description in the present embodiment, reference can be made to side provided by the application embodiment
Method.
Sixth embodiment of the invention is related to a kind of non-volatile memory medium, for storing computer-readable program,
The computer-readable program is used to execute above-mentioned all or part of embodiment of the method for computer.
That is, it will be understood by those skilled in the art that implement the method for the above embodiments be can be with
Relevant hardware is instructed to complete by program, which is stored in a storage medium, including some instructions are to make
It obtains an equipment (can be single-chip microcontroller, chip etc.) or processor (processor) executes side described in each embodiment of the application
The all or part of the steps of method.And storage medium above-mentioned includes: U disk, mobile hard disk, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can be with
Store the medium of program code.
It will be understood by those skilled in the art that the respective embodiments described above are to realize specific embodiments of the present invention,
And in practical applications, can to it, various changes can be made in the form and details, without departing from spirit and model of the invention
It encloses.
The embodiment of the present application discloses a kind of dispatching team load pressure prediction technique of A1., comprising:
Obtain the second load pressure characteristic of the dispatching team for prediction;
According to the second load pressure characteristic and prediction model, the dispatching team is born in preset time
Pressure is carried to be predicted, wherein, the prediction model is the first load pressure characteristic previously according to the dispatching team of acquisition
It is obtained according to training.
A2. dispatching team as described in a1 load pressure prediction technique, the first load pressure of the dispatching team of the acquisition
Power characteristic, specifically:
With predetermined period, the first load pressure characteristic for dispensing team is acquired in preset duration, is obtained
L item the first load pressure characteristic, the L are the amount of cycles for including in the preset duration;First load pressure
Characteristic includes at least team's number, dispenses team in the average load pressure of the past period;
The trained prediction model, specifically:
Using L item the first load pressure characteristic as training set, the training prediction model.
A3. team's load pressure prediction technique is dispensed as described in A2, and the first load pressure characteristic further includes
Following any combination:
Urban information, dispatching team's attribute information, timeslice information, weather pattern, temperature where dispatching team.
A4. dispatching team as described in a1 load pressure prediction technique,
The second load pressure characteristic includes the timeslice to be predicted for specifying the preset time.
A5. dispatching team as described in a1 load pressure prediction technique, it is negative previously according to the first of the dispatching team of acquisition
Carry pressure characterization data, the training prediction model, comprising:
Acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, obtains corresponding to each duration
Dispense the first load pressure characteristic of team;The N is the natural number greater than 1;
According to the first load pressure characteristic of the dispatching team of each duration, respectively to the prediction model into
Row training, the prediction model after obtaining N number of training;
The prediction model after N number of training is tested;
The prediction model after the smallest training of error will be predicted, as target prediction model;
It is described according to the second load pressure characteristic and prediction model, to the dispatching team in preset time
Load pressure predicted, specifically:
According to the second load pressure characteristic and the target prediction model to the dispatching team when default
Between load pressure predicted.
A6. dispatching team as described in a5 load pressure prediction technique, the prediction to after N number of training
Model is tested, and is specifically included:
Obtain N number of test set;The prediction model after N number of test set and N number of training corresponds;
The prediction model after the corresponding training is tested respectively according to N number of test set;
The prediction error of the prediction model after the training according to test result calculations.
A7. the dispatching team load pressure prediction technique as described in A6, the test result obtain in the following manner:
When testing the test data in the test set, the prediction model after the training is exported
Prediction result be compared with the legitimate reading of the test data, if error in the first preset range, determine described in
The test result of test data is correct, otherwise determines the test result mistake of the test data.
A8. the dispatching team load pressure prediction technique as described in A6, the test result obtain in the following manner:
It, will be described in after the training according to preset rules when testing the test data in the test set
The test result specification of prediction model output is 0 or 1;
According to the preset rules by the legitimate reading specification in the test data be 0 or 1;
If the test result after specification is identical as the legitimate reading, the test knot of the test data is determined
Fruit is correct, otherwise determines the test prediction result mistake of the test data.
A9. the dispatching team load pressure prediction technique as described in any one of A1 to A8, the dispatching team of the acquisition
The first load pressure characteristic, be rejecting abnormalities data after dispatching team the first load pressure characteristic:
Wherein, the abnormal data is the load pressure characteristic for meeting the dispatching team of preset condition.
A10. the dispatching team load pressure prediction technique as described in any one of A1 to A8, the class of the prediction model
Type is XGB model.
The embodiment of the present application also discloses a kind of dispatching team load pressure prediction meanss of B1., comprising:
Module is obtained, for obtaining the second load pressure characteristic for being used for the dispatching team of prediction;
Prediction module, for existing to the dispatching team according to the second load pressure characteristic and prediction model
The load pressure of preset time predicted, wherein, the prediction model be previously according to acquisition dispatching team it is first negative
Carry what pressure characterization data training obtained.
The embodiment of the present application also discloses C1. a kind of electronic equipment, including memory and processor, memory storage meter
Calculation machine program, processor execute when running program:
Obtain the second load pressure characteristic of the dispatching team for prediction;
According to the second load pressure characteristic and prediction model, the dispatching team is born in preset time
Pressure is carried to be predicted, wherein, the prediction model is the first load pressure characteristic previously according to the dispatching team of acquisition
It is obtained according to training.
C2. the electronic equipment as described in C1, the first load pressure characteristic of the dispatching team of the acquisition, specifically
Are as follows:
With predetermined period, the first load pressure characteristic for dispensing team is acquired in preset duration, is obtained
L item the first load pressure characteristic, the L are the amount of cycles for including in the preset duration;First load pressure
Characteristic includes at least team's number, dispenses team in the average load pressure of the past period;
The trained prediction model, specifically:
Using L item the first load pressure characteristic as training set, the training prediction model.
C3. the electronic equipment as described in C2, the first load pressure characteristic further includes following any combination:
Urban information, dispatching team's attribute information, timeslice information, weather pattern, temperature where dispatching team.
C4. the electronic equipment as described in C1, the second load pressure characteristic include: timeslice to be predicted;
The second load pressure characteristic includes the timeslice to be predicted for specifying the preset time.
C5. the electronic equipment as described in C1, previously according to acquisition dispatching team the first load pressure characteristic,
The training prediction model, comprising:
Acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, obtains corresponding to each duration
Dispense the first load pressure characteristic of team;The N is the natural number greater than 1;
According to the first load pressure characteristic of the dispatching team of each duration, respectively to the prediction model into
Row training, the prediction model after obtaining N number of training;
The prediction model after N number of training is tested;
The prediction model after the smallest training of error will be predicted, as target prediction model;
It is described according to the second load pressure characteristic and prediction model, to the dispatching team in preset time
Load pressure predicted, specifically:
According to the second load pressure characteristic and the target prediction model to the dispatching team when default
Between load pressure predicted.
C6. the electronic equipment as described in C5, the prediction model to after N number of training are tested, specifically
Include:
Obtain N number of test set;The prediction model after N number of test set and N number of training corresponds;
The prediction model after the corresponding training is tested respectively according to N number of test set;
The prediction error of the prediction model after the training according to test result calculations.
C7. the electronic equipment as described in C6, the test result obtain in the following manner:
When testing the test data in the test set, the prediction model after the training is exported
Prediction result be compared with the legitimate reading of the test data, if error in the first preset range, determine described in
The test result of test data is correct, otherwise determines the test result mistake of the test data.
C8. the electronic equipment as described in C6, the test result obtain in the following manner:
It, will be described in after the training according to preset rules when testing the test data in the test set
The test result specification of prediction model output is 0 or 1;
According to the preset rules by the legitimate reading specification in the test data be 0 or 1;
If the test result after specification is identical as the legitimate reading, the test knot of the test data is determined
Fruit is correct, otherwise determines the test prediction result mistake of the test data.
First load pressure of the C9. electronic equipment as described in any one of C1 to C8, the dispatching team of the acquisition is special
Data are levied, are the first load pressure characteristic of the dispatching team after rejecting abnormalities data:
Wherein, the abnormal data is the load pressure characteristic for meeting the dispatching team of preset condition.
C10. the electronic equipment as described in any one of C1 to C8, the type of the prediction model are XGB model.
The embodiment of the present application also discloses a kind of non-volatile memory medium of D1., for storing computer-readable program,
The dispatching team load pressure that the computer-readable program is used to execute as described in any one of A1 to A10 for computer is pre-
Survey method.
Claims (10)
1. a kind of dispatching team load pressure prediction technique characterized by comprising
Obtain the second load pressure characteristic of the dispatching team for prediction;
According to the second load pressure characteristic and prediction model, to the team that dispenses in the load pressure of preset time
It is predicted, wherein, the prediction model is the first load pressure characteristic training previously according to the dispatching team of acquisition
It obtains.
2. dispatching team according to claim 1 load pressure prediction technique, which is characterized in that the dispatching group of the acquisition
First load pressure characteristic of team, specifically:
With predetermined period, acquire in preset duration the first load pressure characteristic of the dispatching team, L articles obtained the
One load pressure characteristic, the L are the amount of cycles for including in the preset duration;The first load pressure characteristic
According to include at least team number, dispatching team the past period average load pressure;
The trained prediction model, specifically:
Using L item the first load pressure characteristic as training set, the training prediction model.
3. dispatching team according to claim 2 load pressure prediction technique, which is characterized in that first load pressure
Characteristic further includes following any combination:
Urban information, dispatching team's attribute information, timeslice information, weather pattern, temperature where dispatching team.
4. dispatching team according to claim 1 load pressure prediction technique, which is characterized in that previously according to matching for acquisition
Send the first load pressure characteristic of team, the training prediction model, comprising:
Acquisition dispenses the first load pressure characteristic of team respectively in N number of duration, obtains the dispatching for corresponding to each duration
First load pressure characteristic of team;The N is the natural number greater than 1;
According to the first load pressure characteristic of the dispatching team of each duration, the prediction model is instructed respectively
Practice, the prediction model after obtaining N number of training;
The prediction model after N number of training is tested;
The prediction model after the smallest training of error will be predicted, as target prediction model;
It is described according to the second load pressure characteristic and prediction model, to the dispatching team preset time load
Pressure predicted, specifically:
According to the second load pressure characteristic and the target prediction model to the dispatching team in preset time
Load pressure is predicted.
5. dispatching team according to claim 4 load pressure prediction technique, which is characterized in that described to N number of instruction
The prediction model after white silk is tested, and is specifically included:
Obtain N number of test set;The prediction model after N number of test set and N number of training corresponds;
The prediction model after the corresponding training is tested respectively according to N number of test set;
The prediction error of the prediction model after the training according to test result calculations.
6. dispatching team according to claim 5 load pressure prediction technique, which is characterized in that the test result passes through
Following manner obtains:
When testing the test data in the test set, by the prediction of the prediction model output after the training
As a result it is compared with the legitimate reading of the test data, if error in the first preset range, determines the test number
According to test result it is correct, otherwise determine the test result mistake of the test data.
7. dispatching team according to claim 5 load pressure prediction technique, which is characterized in that the test result passes through
Following manner obtains:
When testing the test data in the test set, according to preset rules by the prediction mould after the training
The test result specification of type output is 0 or 1;
According to the preset rules by the legitimate reading specification in the test data be 0 or 1;
If the test result after specification is identical as the legitimate reading, the test result of the test data is being determined just
Really, otherwise determine the test prediction result mistake of the test data.
8. a kind of dispatching team load pressure prediction meanss characterized by comprising
Module is obtained, for obtaining the second load pressure characteristic for being used for the dispatching team of prediction;
Prediction module, for being preset to the dispatching team according to the second load pressure characteristic and prediction model
The load pressure of time predicted, wherein, the prediction model is the first load pressure previously according to the dispatching team of acquisition
The training of power characteristic obtains.
9. a kind of electronic equipment, including memory and processor, memory stores computer program, and processor is held when running program
Row:
Obtain the second load pressure characteristic of the dispatching team for prediction;
According to the second load pressure characteristic and prediction model, to the team that dispenses in the load pressure of preset time
It is predicted, wherein, the prediction model is the first load pressure characteristic training previously according to the dispatching team of acquisition
It obtains.
10. a kind of non-volatile memory medium, for storing computer-readable program, the computer-readable program is by for based on
Calculation machine executes the dispatching team load pressure prediction technique as described in any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811607316.1A CN109685275A (en) | 2018-12-27 | 2018-12-27 | Distribution team load pressure prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811607316.1A CN109685275A (en) | 2018-12-27 | 2018-12-27 | Distribution team load pressure prediction method and device, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109685275A true CN109685275A (en) | 2019-04-26 |
Family
ID=66189922
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811607316.1A Pending CN109685275A (en) | 2018-12-27 | 2018-12-27 | Distribution team load pressure prediction method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109685275A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110378522A (en) * | 2019-07-02 | 2019-10-25 | 北京三快在线科技有限公司 | Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information |
CN111340581A (en) * | 2020-02-11 | 2020-06-26 | 拉扎斯网络科技(上海)有限公司 | Data processing method and device, readable storage medium and electronic equipment |
CN112541246A (en) * | 2019-09-20 | 2021-03-23 | 拉扎斯网络科技(上海)有限公司 | Data processing method and device, readable storage medium and electronic equipment |
CN112799950A (en) * | 2021-02-04 | 2021-05-14 | 北京中交兴路信息科技有限公司 | Background service load capacity test system and method, storage medium and electronic device |
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 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105719008A (en) * | 2015-05-22 | 2016-06-29 | 北京小度信息科技有限公司 | Method and device for performing optimization on delivery system |
CN107092974A (en) * | 2016-11-29 | 2017-08-25 | 北京小度信息科技有限公司 | Dispense pressure prediction method and device |
CN107730087A (en) * | 2017-09-20 | 2018-02-23 | 平安科技(深圳)有限公司 | Forecast model training method, data monitoring method, device, equipment and medium |
CN108197825A (en) * | 2018-01-29 | 2018-06-22 | 北京小度信息科技有限公司 | System scheduling method and device |
CN108427854A (en) * | 2018-03-29 | 2018-08-21 | 深圳市智物联网络有限公司 | A kind of data analysing method and its relevant device |
CN108734390A (en) * | 2018-05-07 | 2018-11-02 | 北京顺丰同城科技有限公司 | A kind of scheduling processing method and device of order dispatching |
-
2018
- 2018-12-27 CN CN201811607316.1A patent/CN109685275A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105719008A (en) * | 2015-05-22 | 2016-06-29 | 北京小度信息科技有限公司 | Method and device for performing optimization on delivery system |
CN107092974A (en) * | 2016-11-29 | 2017-08-25 | 北京小度信息科技有限公司 | Dispense pressure prediction method and device |
CN107730087A (en) * | 2017-09-20 | 2018-02-23 | 平安科技(深圳)有限公司 | Forecast model training method, data monitoring method, device, equipment and medium |
CN108197825A (en) * | 2018-01-29 | 2018-06-22 | 北京小度信息科技有限公司 | System scheduling method and device |
CN108427854A (en) * | 2018-03-29 | 2018-08-21 | 深圳市智物联网络有限公司 | A kind of data analysing method and its relevant device |
CN108734390A (en) * | 2018-05-07 | 2018-11-02 | 北京顺丰同城科技有限公司 | A kind of scheduling processing method and device of order dispatching |
Non-Patent Citations (1)
Title |
---|
携程工程师: "干货,机器学习算法在饿了么供需平衡系统", 《HTTPS://CLOUD.TENCENT.COM/DEVELOPER/ARTICLE/1063420》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110378522A (en) * | 2019-07-02 | 2019-10-25 | 北京三快在线科技有限公司 | Method, apparatus, storage medium and the electronic equipment of prediction dispatching status information |
CN112541246A (en) * | 2019-09-20 | 2021-03-23 | 拉扎斯网络科技(上海)有限公司 | Data processing method and device, 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 |
CN111340581A (en) * | 2020-02-11 | 2020-06-26 | 拉扎斯网络科技(上海)有限公司 | Data processing method and device, readable storage medium and electronic equipment |
CN111340581B (en) * | 2020-02-11 | 2021-01-05 | 拉扎斯网络科技(上海)有限公司 | Data processing method and device, readable storage medium and electronic equipment |
CN112799950A (en) * | 2021-02-04 | 2021-05-14 | 北京中交兴路信息科技有限公司 | Background service load capacity test system and method, storage medium and electronic device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109685275A (en) | Distribution team load pressure prediction method and device, electronic equipment and storage medium | |
Ranis et al. | A theory of economic development | |
CN107908672A (en) | Application report implementation method, equipment and storage medium based on Hadoop platform | |
CN106874483A (en) | A kind of device and method of the patterned quality of data evaluation and test based on big data technology | |
CN110383268A (en) | To the Dynamic Execution of the parametrization application for handling keyed network data flow | |
CN102855525B (en) | A kind of resident's load prediction analytic system and method | |
CN110222739A (en) | Model construction, site cash Demand Forecast method, apparatus and storage medium | |
CN105654338A (en) | Fare by rule calculation method, device and system | |
CN101907696B (en) | Electric energy meter production detecting system and method | |
CN107784105A (en) | Construction of knowledge base method, electronic installation and storage medium based on magnanimity problem | |
CN106651221A (en) | General power enterprise data analysis platform based on state grid index system | |
CN109409005A (en) | DCS design documentation generation method, device, equipment and readable storage medium storing program for executing | |
CN104375943A (en) | Embedded software black-box test case generation method based on static models | |
CN110457751A (en) | A kind of the stock dynamic prediction dispatching method and system of liquefied natural gas receiving station | |
CN106803799A (en) | A kind of performance test methods and device | |
CN110262975A (en) | Test data management method, device, equipment and computer readable storage medium | |
CN110400011A (en) | Output of wind electric field declares the determination method and device of scheme in electric power spot exchange | |
CN104680328A (en) | Power grid construction quality monitoring method based on client perception values | |
CN108921425A (en) | A kind of method, system and the server of asset item classifcation of investment | |
CN108399505A (en) | Distributed photovoltaic power access capacity planing method and terminal device | |
CN116934061A (en) | Block chain-based carbon emission management method, system, equipment and storage medium | |
CN110262961A (en) | Test method, device, storage medium and the terminal device of Workflow Management System | |
CN105260372B (en) | Method based on real-time data base statistics any time period energy consumption | |
CN108960379A (en) | A kind of mobile terminal and system of processing measuring asset dispatching business | |
CN103218745A (en) | Gas station shift completion calculation 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190426 |