CN109543924A - Goods amount prediction technique, device and computer equipment - Google Patents
Goods amount prediction technique, device and computer equipment Download PDFInfo
- Publication number
- CN109543924A CN109543924A CN201811590520.7A CN201811590520A CN109543924A CN 109543924 A CN109543924 A CN 109543924A CN 201811590520 A CN201811590520 A CN 201811590520A CN 109543924 A CN109543924 A CN 109543924A
- Authority
- CN
- China
- Prior art keywords
- goods
- data
- interim data
- time point
- interim
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—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
- G06Q10/083—Shipping
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 present invention provides a kind of goods amount prediction technique, device and computer equipment, which includes: the corresponding history flow of goods interim data of each freightways for obtaining logistics transit depot, is pre-processed to history interim data, generates pretreatment interim data；Using the pretreatment interim data and respective routes information that are in preset first time point as input, pretreatment interim data in preset second time point is as output, goods amount prediction model is trained, goods amount prediction model after being trained, wherein, the second time point was separated by preset duration after first time point and with first time point；Real-time interim data and respective routes information are obtained, the prediction flow of goods interim data by goods amount prediction model after real-time interim data and respective routes information input to training, after obtaining the prefixed time interval of respective routes.Goods amount prediction technique of the invention can reduce the error of goods amount prediction, increase the coverage area of prediction, reduce the cost of prediction.
Description
Technical field
The present invention relates to logistics fields, in particular to a kind of goods amount prediction technique, device, computer equipment and meter
Calculation machine storage medium.
Background technique
It is directed to logistics personnel and the arrangement of logistics vehicles of each route in logistics transit depot at present, needs to carry out in advance
The goods amount of a route is predicted.
The predominantly artificial prediction of the goods amount prediction of current each route, mainly by the staff of a line in logistics transit depot or
Administrative staff predict goods amount, therefore pre- geodesic structure can be influenced by the experience of prognosticator, and brought error is very
Greatly, once prediction error, it will carry out adverse effect for the logistics transfer work belt in future.And the personal foreseeable route model of institute
Enclose very small, and forecast cost is relatively high.
Summary of the invention
In view of the above problems, the present invention provides the storages of a kind of goods amount prediction technique, device, computer equipment and computer
Medium increases the coverage area of prediction, and can reduce the cost of prediction to reduce the error of goods amount prediction.
To achieve the goals above, the present invention adopts the following technical scheme that:
A kind of goods amount prediction technique, for predicting the goods amount of logistics transit depot, comprising:
The corresponding history flow of goods interim data of each freightways for obtaining logistics transit depot, to the history flow of goods transfer
Data are pre-processed, and pretreatment interim data is generated；
Using the pretreatment interim data and respective routes information that are in preset first time point as input, in default
The second time point pretreatment interim data as output, the goods amount prediction model is trained, goods after being trained
Measure prediction model, wherein second time point is separated by default after the first time point and with the first time point
Duration；
Real-time flow of goods interim data and corresponding route information are obtained, by the real-time flow of goods interim data and accordingly
Route information is input to goods amount prediction model after the training, in the prediction flow of goods after obtaining the prefixed time interval of respective routes
Revolution evidence.
Preferably, the flow of goods interim data includes delivering amount, reaching goods amount and operation goods amount.
Preferably, the goods amount prediction model includes in LASSO model, XGBOOST model and THETA algorithm model
It is at least one.
Preferably, described " the corresponding history flow of goods interim data of each freightways of logistics transit depot to be obtained, to described
History flow of goods interim data is pre-processed, and pretreatment interim data is generated " include:
Obtain the corresponding history flow of goods interim data of each route of the logistics transit depot in logistics data library；
The history flow of goods interim data is started the cleaning processing, the invalid number of the history flow of goods interim data is removed
According to, incomplete data and repeated data, interim data after being cleaned；
Obtain the exceptional data point for deviateing normal value after the cleaning in interim data；
Rejecting processing is carried out to the data in preset time before and after the exceptional data point of interim data after the cleaning,
Obtain the pretreatment interim data.
Preferably, the goods amount prediction technique, further includes:
After the pretreatment interim data and respective routes information input to the training at preset third time point
Goods amount prediction model obtains test flow of goods interim data；
Determine the model of the pretreatment interim data at the test flow of goods interim data and preset 4th time point
Deviation, wherein the 4th time point is separated by preset duration after the third time point and with the third time point；
Goods amount prediction model after the training is adjusted according to the model bias.
The present invention also provides a kind of goods amount prediction meanss, for predicting the goods amount of logistics transit depot, comprising:
Data preprocessing module, revolution in the corresponding history flow of goods of each freightways for obtaining logistics transit depot
According to, the history flow of goods interim data is pre-processed, generate pretreatment interim data；
Prediction model training module, for the pretreatment interim data and respective routes of preset first time point will to be in
Information is as input, and the pretreatment interim data in preset second time point is as output, to the goods amount prediction model
Be trained, goods amount prediction model after being trained, wherein second time point after the first time point and with institute
It states first time point and is separated by preset duration；
Goods amount prediction module, for obtaining real-time flow of goods interim data and corresponding route information, by the real-time goods
Goods amount prediction model after stream interim data and respective routes information input to the training, obtains the preset time of respective routes
Prediction flow of goods interim data behind interval.
Preferably, the data preprocessing module includes:
Interim data acquiring unit, each route for obtaining the logistics transit depot in logistics data library is corresponding described to be gone through
History flow of goods interim data；
Interim data cleaning unit removes the history for starting the cleaning processing to the history flow of goods interim data
Invalid data, incomplete data and the repeated data of flow of goods interim data, interim data after being cleaned；
Abnormal data acquiring unit, for obtaining the exceptional data point for deviateing normal value after the cleaning in interim data；
Abnormal data elimination unit, for preset time before and after the exceptional data point to interim data after the cleaning
Interior data carry out rejecting processing, obtain the pretreatment interim data.
Preferably, the goods amount prediction meanss, further includes:
Test data obtains module, for by the pretreatment interim data and respective routes at preset third time point
Goods amount prediction model after information input to the training obtains test flow of goods interim data；
Model bias determining module, for determining described in the test flow of goods interim data and preset 4th time point
Pre-process interim data model bias, wherein the 4th time point after the third time point and with the third
Time point is separated by preset duration；
Prediction model adjusts module, for being adjusted according to the model bias to goods amount prediction model after the training
It is whole.
The present invention also provides a kind of computer equipments, including memory and processor, and the memory is based on storing
Calculation machine program, the processor runs the computer program so that the computer equipment executes the goods amount prediction side
Method.
The present invention also provides a kind of computer storage medium, it is stored with calculating used in the computer equipment
Machine program.
The present invention provides a kind of goods amount prediction technique, which is used to predict the goods amount of logistics transit depot, packet
It includes: the corresponding history flow of goods interim data of each freightways of logistics transit depot is obtained, to the history flow of goods interim data
It is pre-processed, generates pretreatment interim data；By the pretreatment interim data for being in preset first time point and corresponding road
For line information as input, the pretreatment interim data in preset second time point predicts mould as output, to the goods amount
Type is trained, goods amount prediction model after being trained, wherein second time point after the first time point and with
The first time point is separated by preset duration；Real-time flow of goods interim data and corresponding route information are obtained, by the reality
When flow of goods interim data and respective routes information input to the training after goods amount prediction model, obtain the default of respective routes
Prediction flow of goods interim data after time interval.Goods amount prediction technique of the invention is carried out in logistics using machine learning method
The prediction of transition flow of goods interim data can reduce the error of goods amount prediction, increase the coverage area of prediction, and can reduce
The cost of prediction.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of the scope of the invention.
Fig. 1 is a kind of flow chart for goods amount prediction technique that the embodiment of the present invention 1 provides；
Fig. 2 is a kind of flow chart of the data prediction for goods amount prediction technique that the embodiment of the present invention 2 provides；
Fig. 3 is a kind of flow chart for goods amount prediction technique that the embodiment of the present invention 3 provides；
Fig. 4 is a kind of structural schematic diagram for goods amount prediction meanss that the embodiment of the present invention 4 provides；
Fig. 5 is a kind of structural schematic diagram of the data preprocessing module for goods amount prediction meanss that the embodiment of the present invention 4 provides；
Fig. 6 is the structural schematic diagram for another goods amount prediction meanss that the embodiment of the present invention 4 provides.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
Embodiment 1
Fig. 1 is a kind of flow chart for goods amount prediction technique that the embodiment of the present invention 1 provides, and this method comprises the following steps:
Step S11: the corresponding history flow of goods interim data of each freightways of logistics transit depot is obtained, to history flow of goods
Interim data is pre-processed, and pretreatment interim data is generated.
In the embodiment of the present invention, computer equipment can be set, which connects the big data of logistics center
Library, the corresponding history flow of goods interim data of each route that corresponding logistics transit depot is obtained by large database concept.Wherein, in the flow of goods
Revolution refers to that logistics transit depot issues on the route according to including delivering amount, arrival goods amount and operation goods amount, the delivering amount
The quantity of cargo reaches goods amount and refers to that logistics transit depot receives the quantity of cargo on the route, and operates goods amount and then refer to
The quantity for the cargo that logistics transit depot is operated in some time threshold.The delivering amount reaches goods amount and operation goods amount
Connect corresponding and time point, the delivering amount of different time points or period reach goods amount and operation goods amount is all different.
In the embodiment of the present invention, computer equipment obtains the corresponding history flow of goods interim data of each route of logistics transit depot
Afterwards, which will be pre-processed, and will generate pretreatment interim data.Wherein, the preprocessing process is main
The data for situations such as history flow of goods interim data is arranged, invalid, repetition and missing are removed, so as to the pretreatment of generation
Interim data is good training sample, to improve the precision to the training of goods amount prediction model.The preprocessing process can be with
It is realized by algorithm or application program, such as data prediction application program can be provided in computer equipment, obtained
After taking history flow of goods interim data, it can use the application program and the history flow of goods interim data pre-processed, obtain pre-
Handle interim data.
Step S12: using in preset first time point pretreatment interim data and respective routes information as input,
Pretreatment interim data in preset second time point is trained goods amount prediction model, is trained as output
Goods amount prediction model afterwards, wherein the second time point was separated by preset duration after first time point and with first time point.
Amount prediction model available in stock can be pre-established in the embodiment of the present invention, in the computer equipment, wherein the goods amount is pre-
Survey model include LASSO model (LASSO, least absolute shrinkage and selection operator, most
The convergence of small absolute value and selection operator), XGBOOST model and THETA algorithm model.Interim data is pre-processed to this utilizing
When goods amount prediction model is trained, route information and the pretreatment interim data of the route first time point can be made
Pretreatment interim data for input, the second time point is trained as output, and second time point is in first time point
After prefixed time interval namely the pretreatment interim data at the second time point generated in the future of first time point, therefore instructed
Goods amount prediction model after white silk has the ability of prediction.Wherein, the prefixed time interval can be seven days, ten days or one month,
It is not limited here.The pretreatment interim data, the pretreatment interim data and correspondence at the second time point of the first time point
Route information be goods amount prediction model training sample, the training sample will by pretreatment interim data utilize time sampling
Mode extract, therefore a large amount of training sample can be obtained.Sampling acquisition is being carried out by pre-processing interim data
Training sample can also be used as the test sample of goods amount prediction model after later period training.
Step S13: obtaining real-time flow of goods interim data and corresponding route information, by real-time flow of goods interim data and
Goods amount prediction model after respective routes information input to training, in the prediction flow of goods after obtaining the prefixed time interval of respective routes
Revolution evidence.
In the embodiment of the present invention, when carrying out the prediction of goods amount, the available real-time flow of goods interim data of the computer equipment
And corresponding route information, and by the real-time flow of goods interim data and corresponding route data be input to training after goods amount it is pre-
Model is surveyed, thus the prediction flow of goods interim data after obtaining the prefixed time intervals of respective routes.Wherein, the real-time flow of goods transfer
Data and respective routes can be input in the computer equipment by staff, carried out the data of goods amount, do not limited here
It is fixed.
Embodiment 2
Fig. 2 is a kind of flow chart of the data prediction for goods amount prediction technique that the embodiment of the present invention 2 provides, including as follows
Step:
Step S21: the corresponding history flow of goods interim data of each route of the logistics transit depot in logistics data library is obtained.
Step S22: starting the cleaning processing history flow of goods interim data, removes the invalid number of history flow of goods interim data
According to, incomplete data and repeated data, interim data after being cleaned.
In the embodiment of the present invention, which, can be with when obtaining the history flow of goods interim data in logistics data library
The interim data of a period of time is obtained, or in the way of sampling, obtains the flow of goods interim data of daily set time.It is counting
After calculating machine equipment acquisition history flow of goods interim data, which can be pre-processed, it first can be right
The history flow of goods interim data carries out data cleansing operation, removes invalid data, incomplete data in history flow of goods interim data
And repeated data etc., the interim data after being cleaned.
It is above-mentioned that algorithm can use to history flow of goods interim data progress data cleansing treatment process in the embodiment of the present invention
Or application program is realized, such as can be provided with data cleansing application program in computer equipment, the computer equipment from
After obtaining history flow of goods interim data in flow of goods database, the data cleansing application program can use to revolution in history flow of goods
According to starting the cleaning processing, interim data after being cleaned.
Step S23: the exceptional data point for deviateing normal value after cleaning in interim data is obtained.
In the embodiment of the present invention, after computer equipment acquisition cleaning after interim data, transfer after the cleaning can also be obtained
Deviate the exceptional data point of normal value in data.For example, available interim data is changed greatly compared to the interim data of front and back
Point, be exceptional data point, wherein the variation determine threshold value can there is staff to be defined.For example, double 11
When, the goods amount of the interim data is obviously more much higher than the goods amount of usual interim data, so that it may determine double 11 this when
Between point be exceptional data point.
In the embodiment of the present invention, the process of the acquisition exceptional data point can use algorithm or application program to realize, example
The application program of search exceptional data point can be such as provided in computer equipment, in after application program search cleaning
Revolution front and back variation in is more than the exceptional data point of preset threshold.
Step S24: carrying out rejecting processing to the data in preset time before and after the exceptional data point of interim data after cleaning,
Obtain pretreatment interim data.
In the embodiment of the present invention, computer equipment obtain cleaning after interim data exceptional data point after, can basis
Preset time carries out rejecting processing to the front and back data of the abnormal point, obtains pretreatment interim data.It also i.e. will be where abnormal point
One piece of data, abnormal point is removed in a manner of the data for rejecting preset time period, generate preprocessed data, avoid abnormal point
The training of goods amount prediction model is affected.
Embodiment 3
Fig. 3 is a kind of flow chart for goods amount prediction technique that the embodiment of the present invention 3 provides, and this method comprises the following steps:
Step S31: the corresponding history flow of goods interim data of each freightways of logistics transit depot is obtained, to history flow of goods
Interim data is pre-processed, and pretreatment interim data is generated.
This step is consistent with above-mentioned steps S11, and details are not described herein.
Step S32: using in preset first time point pretreatment interim data and respective routes information as input,
Pretreatment interim data in preset second time point is trained goods amount prediction model, is trained as output
Goods amount prediction model afterwards, wherein the second time point was separated by preset duration after first time point and with first time point.
This step is consistent with above-mentioned steps S12, and details are not described herein.
Step S33: obtaining real-time flow of goods interim data and corresponding route information, by real-time flow of goods interim data and
Goods amount prediction model after respective routes information input to training, in the prediction flow of goods after obtaining the prefixed time interval of respective routes
Revolution evidence.
This step is consistent with above-mentioned steps S13, and details are not described herein.
Step S34: after the pretreatment interim data at preset third time point and respective routes information input to training
Goods amount prediction model obtains test flow of goods interim data.
It, can also be from pre- outlet transfer after the trained goods amount prediction model of computer equipment in the embodiment of the present invention
The pretreatment interim data of specified time point and the goods amount prediction after respective routes information input to the training are chosen in data
In model, the goods amount prediction model after the training is tested.Wherein, which selects in pretreatment interim data
The pretreatment interim data at third time point out, and by the pretreatment interim data and respective routes information at the third time point
It is input to goods amount prediction model after training, obtains the test volume of goods flow interim data of model output.
Step S35: the model of the pretreatment interim data at test flow of goods interim data and preset 4th time point is determined
Deviation, wherein the 4th time point was separated by preset duration after third time point and with third time point.
In the embodiment of the present invention, computer equipment will utilize the test flow of goods after obtaining test volume of goods flow interim data
Interim data is measured compared with the pretreatment interim data at the 4th time point carries out matching, wherein the 4th time point is in third
Between after the prefixed time interval put namely the pretreatment interim data at the 4th time point is goods amount prediction model after the training
The normal data that should be exported.After data are matched relatively, the model that can obtain goods amount prediction model after current training is inclined
Difference.Wherein, which can use algorithm or application program to realize, here without limitation.
Step S36: goods amount prediction model after training is adjusted according to model bias.
In the embodiment of the present invention, the model bias of goods amount prediction model after computer equipment obtains training by test
Afterwards, which can be adjusted goods amount prediction model according to model bias, correction model deviation.Wherein, the survey
Examination can be repeated constantly with adjustment process, and the equal difference for the pretreatment interim data that test is chosen every time, to be continuously improved
The precision of goods amount prediction model.
Embodiment 4
Fig. 4 is a kind of structural schematic diagram for goods amount prediction meanss that the embodiment of the present invention 4 provides.
The goods amount prediction meanss 400 include:
Data preprocessing module 410, for obtaining the corresponding history flow of goods transfer of each freightways of logistics transit depot
Data pre-process the history flow of goods interim data, generate pretreatment interim data.
Prediction model training module 420, for the pretreatment interim data of preset first time point and corresponding will to be in
Route information as input, predict the goods amount as output by the pretreatment interim data in preset second time point
Model is trained, goods amount prediction model after being trained, wherein second time point after the first time point simultaneously
It is separated by preset duration with the first time point.
Goods amount prediction module 430 will be described real-time for obtaining real-time flow of goods interim data and corresponding route information
Goods amount prediction model after flow of goods interim data and respective routes information input to the training, obtain respective routes it is default when
Between be spaced after prediction flow of goods interim data.
As shown in figure 5, the data preprocessing module 410 includes:
Interim data acquiring unit 411, the corresponding institute of each route for obtaining the logistics transit depot in logistics data library
State history flow of goods interim data.
Interim data cleaning unit 412 removes in history flow of goods for starting the cleaning processing to history flow of goods interim data
Invalid data, incomplete data and the repeated data of revolution evidence, interim data after being cleaned.
Abnormal data acquiring unit 413, for obtaining the exceptional data point for deviateing normal value after cleaning in interim data.
Abnormal data elimination unit 414, in preset time before and after the exceptional data point to interim data after cleaning
Data carry out rejecting processing, obtain pretreatment interim data.
As shown in fig. 6, the goods amount prediction meanss 400 further include:
Test data obtains module 440, for by the pretreatment interim data at preset third time point and accordingly
Route information is input to goods amount prediction model after the training, obtains test flow of goods interim data.
Model bias determining module 450, for determining the test flow of goods interim data and preset 4th time point
It is described pretreatment interim data model bias, wherein the 4th time point after the third time point and with it is described
Third time point is separated by preset duration.
Prediction model adjusts module 460, for being adjusted according to model bias to goods amount prediction model after training.
In the embodiment of the present invention, the above-mentioned more detailed function description of modules can be with reference to corresponding in previous embodiment
Partial content, details are not described herein.
In addition, the computer equipment includes memory and processor, storage the present invention also provides a kind of computer equipment
Device can be used for storing computer program, and processor is by running the computer program, so that it is above-mentioned to execute computer equipment
The function of method or the modules in above-mentioned goods amount prediction meanss.
Memory may include storing program area and storage data area, wherein storing program area can storage program area, at least
Application program needed for one function (such as sound-playing function, image player function etc.) etc.；Storage data area can store root
Created data (such as audio data, phone directory etc.) etc. are used according to computer equipment.In addition, memory may include height
Fast random access memory, can also include nonvolatile memory, a for example, at least disk memory, flush memory device,
Or other volatile solid-state parts.
The present embodiment additionally provides a kind of computer storage medium, for storing calculating used in above-mentioned computer equipment
Machine program.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and structure in attached drawing
Figure shows the system frame in the cards of the device of multiple embodiments according to the present invention, method and computer program product
Structure, function and operation.In this regard, each box in flowchart or block diagram can represent a module, section or code
A part, a part of the module, section or code includes one or more for implementing the specified logical function
Executable instruction.It should also be noted that function marked in the box can also be to be different from the implementation as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that in structure chart and/or flow chart
The combination of each box and the box in structure chart and/or flow chart, can function or movement as defined in executing it is dedicated
Hardware based system realize, or can realize using a combination of dedicated hardware and computer instructions.
In addition, each functional module or unit in each embodiment of the present invention can integrate one independence of formation together
Part, be also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be intelligence
Can mobile phone, personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory),
Random access memory (RAM, Random Access Memory), magnetic or disk etc. be various to can store program code
Medium.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of goods amount prediction technique, for predicting the goods amount of logistics transit depot characterized by comprising
The corresponding history flow of goods interim data of each freightways for obtaining logistics transit depot, to the history flow of goods interim data
It is pre-processed, generates pretreatment interim data；
Using in preset first time point pretreatment interim data and respective routes information as input, be in preset the
The pretreatment interim data at two time points is trained the goods amount prediction model, goods amount is pre- after being trained as output
Survey model, wherein second time point is after the first time point and when being separated by preset with the first time point
It is long；
Real-time flow of goods interim data and corresponding route information are obtained, by the real-time flow of goods interim data and respective routes
Goods amount prediction model after information input to the training, revolution in the prediction flow of goods after obtaining the prefixed time interval of respective routes
According to.
2. goods amount prediction technique according to claim 1, which is characterized in that the flow of goods interim data includes to deliver
Amount reaches goods amount and operation goods amount.
3. goods amount prediction technique according to claim 1, which is characterized in that the goods amount prediction model includes LASSO mould
At least one of type, XGBOOST model and THETA algorithm model.
4. goods amount prediction technique according to claim 1, which is characterized in that described " to obtain each goods of logistics transit depot
The corresponding history flow of goods interim data of route is transported, the history flow of goods interim data is pre-processed, generates pretreatment transfer
Data " include:
Obtain the corresponding history flow of goods interim data of each route of the logistics transit depot in logistics data library；
The history flow of goods interim data is started the cleaning processing, the invalid data, residual of the history flow of goods interim data is removed
Lack data and repeated data, interim data after being cleaned；
Obtain the exceptional data point for deviateing normal value after the cleaning in interim data；
Rejecting processing is carried out to the data in preset time before and after the exceptional data point of interim data after the cleaning, is obtained
The pretreatment interim data.
5. goods amount prediction technique according to claim 1, which is characterized in that further include:
By goods amount after the pretreatment interim data and respective routes information input to the training at preset third time point
Prediction model obtains test flow of goods interim data；
Determine the model bias of the pretreatment interim data at the test flow of goods interim data and preset 4th time point,
Wherein, the 4th time point is separated by preset duration after the third time point and with the third time point；
Goods amount prediction model after the training is adjusted according to the model bias.
6. a kind of goods amount prediction meanss, for predicting the goods amount of logistics transit depot characterized by comprising
Data preprocessing module is right for obtaining the corresponding history flow of goods interim data of each freightways of logistics transit depot
The history flow of goods interim data is pre-processed, and pretreatment interim data is generated；
Prediction model training module, for the pretreatment interim data and respective routes information of preset first time point will to be in
As input, the pretreatment interim data in preset second time point carries out the goods amount prediction model as output
Training, goods amount prediction model after being trained, wherein second time point is after the first time point and with described the
One time point was separated by preset duration；
Goods amount prediction module will be in the real-time flow of goods for obtaining real-time flow of goods interim data and corresponding route information
Revolution goods amount prediction model accordingly and after respective routes information input to the training, obtains the prefixed time interval of respective routes
Prediction flow of goods interim data afterwards.
7. goods amount prediction meanss according to claim 6, which is characterized in that the data preprocessing module includes:
Interim data acquiring unit, the corresponding history goods of each route for obtaining the logistics transit depot in logistics data library
Flow interim data；
Interim data cleaning unit removes the history flow of goods for starting the cleaning processing to the history flow of goods interim data
Invalid data, incomplete data and the repeated data of interim data, interim data after being cleaned；
Abnormal data acquiring unit, for obtaining the exceptional data point for deviateing normal value after the cleaning in interim data；
Abnormal data elimination unit, in preset time before and after the exceptional data point to interim data after the cleaning
Data carry out rejecting processing, obtain the pretreatment interim data.
8. goods amount prediction meanss according to claim 6, which is characterized in that further include:
Test data obtains module, for by the pretreatment interim data and respective routes information at preset third time point
It is input to goods amount prediction model after the training, obtains test flow of goods interim data；
Model bias determining module, for determining the pre- place of test the flow of goods interim data and preset 4th time point
Manage interim data model bias, wherein the 4th time point after the third time point and with the third time
Point is separated by preset duration；
Prediction model adjusts module, for being adjusted according to the model bias to goods amount prediction model after the training.
9. a kind of computer equipment, which is characterized in that including memory and processor, the memory is for storing computer
Program, the processor runs the computer program so that the computer equipment executes according to claim 1 to any in 5
Goods amount prediction technique described in.
10. a kind of computer storage medium, which is characterized in that it, which is stored in computer equipment as claimed in claim 9, is made
Computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811590520.7A CN109543924A (en) | 2018-12-20 | 2018-12-20 | Goods amount prediction technique, device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811590520.7A CN109543924A (en) | 2018-12-20 | 2018-12-20 | Goods amount prediction technique, device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109543924A true CN109543924A (en) | 2019-03-29 |
Family
ID=65858167
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811590520.7A Pending CN109543924A (en) | 2018-12-20 | 2018-12-20 | Goods amount prediction technique, device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109543924A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993374A (en) * | 2019-04-15 | 2019-07-09 | 成都四方伟业软件股份有限公司 | Car loading prediction technique and device |
| CN110866786A (en) * | 2019-11-12 | 2020-03-06 | 德邦物流股份有限公司 | Goods quantity prediction method and device, electronic equipment and storage medium |
| CN111325398A (en) * | 2020-02-19 | 2020-06-23 | 上海东普信息科技有限公司 | Method and system for predicting quantity of goods in network, and storage medium |
| CN111340278A (en) * | 2020-02-19 | 2020-06-26 | 上海东普信息科技有限公司 | Method for predicting destination cargo volume and storage medium |
| CN111507542A (en) * | 2020-04-30 | 2020-08-07 | 南京福佑在线电子商务有限公司 | Goods quantity measuring method and device and electronic equipment |
| CN112131793A (en) * | 2020-09-24 | 2020-12-25 | 上海燕汐软件信息科技有限公司 | Method and device for simulating line cargo volume |
| CN113034090A (en) * | 2021-05-26 | 2021-06-25 | 北京每日优鲜电子商务有限公司 | Transportation equipment scheduling method and device, electronic equipment and computer readable medium |
| CN113077072A (en) * | 2020-01-06 | 2021-07-06 | 北京京邦达贸易有限公司 | Method and device for predicting cargo quantity |
| CN114154709A (en) * | 2021-11-30 | 2022-03-08 | 北京京东振世信息技术有限公司 | Method, device, equipment and medium for predicting line transportation cargo quantity |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103310286A (en) * | 2013-06-25 | 2013-09-18 | 浙江大学 | Product order prediction method and device with time series characteristics |
| CN105976153A (en) * | 2016-06-20 | 2016-09-28 | 中联物流（中国）有限公司 | Vehicle departure data predication method of logistics management system |
| WO2017056367A1 (en) * | 2015-09-30 | 2017-04-06 | 日本電気株式会社 | Information processing system, information processing method, and information processing program |
| CN107944598A (en) * | 2017-10-31 | 2018-04-20 | 苏宁云商集团股份有限公司 | A kind of logistics route collocation method and system |
| CN108564326A (en) * | 2018-04-19 | 2018-09-21 | 安吉汽车物流股份有限公司 | Prediction technique and device, computer-readable medium, the logistics system of order |
-
2018
- 2018-12-20 CN CN201811590520.7A patent/CN109543924A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103310286A (en) * | 2013-06-25 | 2013-09-18 | 浙江大学 | Product order prediction method and device with time series characteristics |
| WO2017056367A1 (en) * | 2015-09-30 | 2017-04-06 | 日本電気株式会社 | Information processing system, information processing method, and information processing program |
| CN105976153A (en) * | 2016-06-20 | 2016-09-28 | 中联物流（中国）有限公司 | Vehicle departure data predication method of logistics management system |
| CN107944598A (en) * | 2017-10-31 | 2018-04-20 | 苏宁云商集团股份有限公司 | A kind of logistics route collocation method and system |
| CN108564326A (en) * | 2018-04-19 | 2018-09-21 | 安吉汽车物流股份有限公司 | Prediction technique and device, computer-readable medium, the logistics system of order |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993374B (en) * | 2019-04-15 | 2021-01-05 | 成都四方伟业软件股份有限公司 | Cargo quantity prediction method and device |
| CN109993374A (en) * | 2019-04-15 | 2019-07-09 | 成都四方伟业软件股份有限公司 | Car loading prediction technique and device |
| CN110866786A (en) * | 2019-11-12 | 2020-03-06 | 德邦物流股份有限公司 | Goods quantity prediction method and device, electronic equipment and storage medium |
| CN113077072B (en) * | 2020-01-06 | 2024-06-18 | 北京京邦达贸易有限公司 | Method and device for predicting cargo quantity |
| CN113077072A (en) * | 2020-01-06 | 2021-07-06 | 北京京邦达贸易有限公司 | Method and device for predicting cargo quantity |
| CN111340278A (en) * | 2020-02-19 | 2020-06-26 | 上海东普信息科技有限公司 | Method for predicting destination cargo volume and storage medium |
| CN111340278B (en) * | 2020-02-19 | 2023-10-24 | 上海东普信息科技有限公司 | Destination cargo quantity prediction method and storage medium |
| CN111325398A (en) * | 2020-02-19 | 2020-06-23 | 上海东普信息科技有限公司 | Method and system for predicting quantity of goods in network, and storage medium |
| CN111507542A (en) * | 2020-04-30 | 2020-08-07 | 南京福佑在线电子商务有限公司 | Goods quantity measuring method and device and electronic equipment |
| CN111507542B (en) * | 2020-04-30 | 2021-08-17 | 南京福佑在线电子商务有限公司 | Freight order quantity measuring method and device and electronic equipment |
| CN112131793A (en) * | 2020-09-24 | 2020-12-25 | 上海燕汐软件信息科技有限公司 | Method and device for simulating line cargo volume |
| CN112131793B (en) * | 2020-09-24 | 2023-08-04 | 上海燕汐软件信息科技有限公司 | Line cargo quantity simulation method and device |
| CN113034090A (en) * | 2021-05-26 | 2021-06-25 | 北京每日优鲜电子商务有限公司 | Transportation equipment scheduling method and device, electronic equipment and computer readable medium |
| CN113034090B (en) * | 2021-05-26 | 2021-09-03 | 北京每日优鲜电子商务有限公司 | Transportation equipment scheduling method and device, electronic equipment and computer readable medium |
| CN114154709A (en) * | 2021-11-30 | 2022-03-08 | 北京京东振世信息技术有限公司 | Method, device, equipment and medium for predicting line transportation cargo quantity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109543924A (en) | Goods amount prediction technique, device and computer equipment | |
| EP3514823A2 (en) | Machine learning-based semiconductor manufacturing yield prediction system and method | |
| EP3422518B1 (en) | A method for recognizing contingencies in a power supply network | |
| CN109272509A (en) | A kind of object detection method of consecutive image, device, equipment and storage medium | |
| KR102335180B1 (en) | Apparatus and method for estimating aging condition of battery | |
| Djenouri et al. | Fast and accurate deep learning framework for secure fault diagnosis in the industrial internet of things | |
| CN111310518B (en) | Picture feature extraction method, target re-identification method, device and electronic equipment | |
| CN110210304A (en) | Method and system for target detection and tracking | |
| CN111291170B (en) | Session recommendation method and related device based on intelligent customer service | |
| CN110097320A (en) | Order forecast method and device, logistics system and computer-readable medium | |
| EP3422517A1 (en) | A method for recognizing contingencies in a power supply network | |
| CN111494964B (en) | Virtual article recommendation method, model training method, device and storage medium | |
| CN108885628A (en) | Data analysing method candidate's determination device | |
| CN109740567A (en) | Key point location model training method, localization method, device and equipment | |
| US10824956B1 (en) | System and method for price estimation of reports before execution in analytics | |
| Wimmer et al. | Dimensionality reduced training by pruning and freezing parts of a deep neural network: a survey | |
| TW202327357A (en) | Video processing using delta distillation | |
| CN108306997A (en) | Domain name mapping monitoring method and device | |
| CN113392867A (en) | Image identification method and device, computer equipment and storage medium | |
| CN111160594B (en) | Method and device for estimating arrival time and storage medium | |
| WO2016015170A1 (en) | A method for face recognition and a system thereof | |
| CN111222370A (en) | Case studying and judging method, system and device | |
| CN116128044A (en) | Model pruning method, image processing method and related devices | |
| Davila-Frias et al. | Deep neural networks for all-terminal network reliability estimation | |
| CN110084455A (en) | A kind of data processing method, apparatus and system |
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: 20190329 |