CN109978208A - A kind of time series data prediction technique, device and computer storage medium - Google Patents
A kind of time series data prediction technique, device and computer storage medium Download PDFInfo
- Publication number
- CN109978208A CN109978208A CN201711456483.6A CN201711456483A CN109978208A CN 109978208 A CN109978208 A CN 109978208A CN 201711456483 A CN201711456483 A CN 201711456483A CN 109978208 A CN109978208 A CN 109978208A
- Authority
- CN
- China
- Prior art keywords
- amount
- part amount
- historical data
- time
- impact
- 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 25
- 230000005856 abnormality Effects 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 13
- 238000009825 accumulation Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 230000007423 decrease Effects 0.000 claims description 6
- 230000001186 cumulative effect Effects 0.000 claims description 2
- 230000007774 longterm Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 3
- 238000007418 data mining Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 239000000654 additive Substances 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 2
- 239000012491 analyte Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 125000003367 polycyclic group Chemical group 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0838—Historical data
Abstract
This application discloses a kind of time series data prediction technique, device and computer storage mediums, wherein this method comprises: obtaining the historical data for meeting setting rule in logistic industry in the first setting duration；Based on the historical data, the prediction model for predicting timing is established；Based on the prediction model, the part amount in logistic industry is predicted, is able to carry out and is effectively predicted, and then meets loglstics enterprise to the demand of part amount predicted steadily in the long term.
Description
Technical field
The present invention relates generally to technical field of data processing, and in particular to time series data prediction technique, device and calculating
Machine storage medium.
Background technique
Prediction is big data core value, and the time series data how to generate to the historical data of magnanimity and in real time carries out quick
, accurately data mining, to provide the prediction data of high quality, be academic circles at present and industry research hotspot it
One.
State of the Time Series Data Mining using things in different moments is formed by data as research object, passes through clock synchronization
Between the feature of sequence data analyzed and studied, the development and change rule of things is disclosed, for instructing society, the warp of people
Ji, military and life etc. are movable.Time series excavation be of great significance to human society, science and technology and expanding economy, just by
Gradually become one of the research hotspot of data mining.For logistic industry, time series data prediction is mainly using traditional timing mould
Three kinds of type, Generalized Additive Models and tree-model methods.Common practice is that the history receipts after input cleaning send number of packages evidence, is then analyzed
Its trend that changes with time establishes model to carry out outside forecast.
But due to the movable particularity of logistic industry, model above can not be effectively predicted, and then be unable to satisfy object
Enterprise is flowed to the demand of part amount predicted steadily in the long term.
Summary of the invention
It in view of drawbacks described above in the prior art or deficiency, is intended to provide one kind and is able to carry out and be effectively predicted, and then meet
Method of the loglstics enterprise to the demand of part amount predicted steadily in the long term.
A kind of time series data prediction technique, which comprises obtain in logistic industry and meet in the first setting duration
Set the historical data of rule；Based on the historical data, the prediction model for predicting timing is established；Based on the prediction mould
Type predicts the part amount in logistic industry.
The historical data includes to meet the part amount of setting rule in logistic industry in the first setting duration；Based on described
Historical data establishes the prediction model for predicting timing, comprising: is based on the historical data, respectively the growth of determining member amount
The impact of trend, the mechanical periodicity of part amount and particular time to part amount, wherein the particular time is that finger amount is flown up
Or the period declined suddenly；Add up the impact of the growth trend, mechanical periodicity and particular time to part amount；What is obtained is cumulative
As a result the prediction model as part amount.
Further include: determine that lag item, the lag item refer to the explained variable of phase in the past to current explained variable
It has an impact；Based on the historical data, the prediction model for predicting timing is established, comprising: it is based on the historical data, point
The impact to part amount of mechanical periodicity, particular time of the growth trend of other determining member amount, part amount, and lag item, wherein described
Particular time is the period that finger amount flies up or declines suddenly；Add up the growth trend, mechanical periodicity, particular time pair
Send impact and the lag item of part amount；Obtained accumulation result is as the prediction model for sending part amount.
Determine impact of the particular time to part amount, comprising: determine the time of part amount variation abnormality in the historical data,
Described in variation abnormality be the first setting value to be increased beyond in finger amount curve or the situation of the second setting value was looked into decline；By institute
The time for stating variation abnormality is divided into different time windows；According to the time window, determine that particular time rushes part amount
It hits.
The part amount includes the addressee amount of logistic industry；Or logistic industry sends part amount.
A kind of time series data prediction meanss, described device include: acquisition module, are set for obtaining in logistic industry first
Meet the historical data of setting rule in timing is long；Processing module is established for being based on the historical data for predicting timing
Prediction model；And it is based on the prediction model, the part amount in logistic industry is predicted.
The historical data includes to meet the part amount of setting rule in logistic industry in the first setting duration；The processing
Module is specifically used for being based on the historical data, respectively the growth trend of determining member amount, the mechanical periodicity of part amount and special
Impact of the period to part amount, wherein the particular time is the period that finger amount flies up or declines suddenly；Add up the increasing
The impact of long trend, mechanical periodicity and particular time to part amount；Prediction model of the obtained accumulation result as part amount.
The acquisition module is also used to determine that lag item, the lag item refer to the explained variable of phase in the past to current
Explained variable have an impact；The processing module is also used to based on the historical data, and the growth of determining member amount becomes respectively
The impact of gesture, the mechanical periodicity of part amount, particular time to part amount, and lag item, wherein the particular time is that finger amount is prominent
The period for so rising or declining suddenly；The growth trend, mechanical periodicity, particular time add up to the impact of sending part amount and stagnant
It is consequent；Obtained accumulation result is as the prediction model for sending part amount.
The processing module, specifically for the time of part amount variation abnormality in the determination historical data, wherein the change
Changing abnormal is the first setting value to be increased beyond in finger amount curve or the situation of the second setting value was looked into decline；The variation is different
The normal time is divided into different time windows；According to the time window, impact of the particular time to part amount is determined.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
The step of time series data prediction technique is realized when row.
By using above-mentioned technical proposal, it is based on historical data, establishes prediction model, then the prediction mould by establishing
Type predicts the part amount in logistic industry solve in the prior art due to the movable particularity of logistic industry, with upper mold
Type can not be effectively predicted, and then the problem of be unable to satisfy the demand steadily in the long term predicted of the loglstics enterprise to part amount.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 shows a kind of one of time series data prediction technique flow chart；
Fig. 2 shows the two of a kind of time series data prediction technique flow chart；
Fig. 3 shows a kind of time series data prediction meanss structure composition schematic diagram.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Referring to FIG. 1, it illustrates the illustrative methods that can apply the embodiment of the present application.
Step 11, historical data is obtained.
In the technical solution that the embodiment of the present invention proposes, historical data refers to the part amount data stored in logistic industry.
Part amount includes addressee amount, also may include outbox amount.In the database, either on line or line under, can store send part amount,
The information of addressee amount.Those information can be, but not limited to: the coding of part amount, time.Time, which can be, daily to be stored, can also
Being stored according to the moon, can also be stored according to the specific time of input system.
Step 12, the historical data of acquisition is pre-processed.
The historical data of acquisition is pre-processed, unwanted information in the historical data of acquisition is removed.
In the technical solution that the embodiment of the present invention proposes, predicted primarily directed to part amount, therefore can remove and go through
The odd numbers information for including in history data.
Specifically, in the technical solution that the embodiment of the present invention proposes, historical data includes addressee amount and/or outbox amount,
Each site addressee amount (having order, without order) can be transferred from database according to different business scenarios and sends the letter of part amount
Breath, below will be using peak command scheduling addressee amount as test data, and the date where historical data is 2015/05/01-2017/07/
28, extrapolation is described in detail for being predicted within 31 days.The historical data of acquisition can be such as following table 1 institute after pretreatment
Show.
Table 1
| Coding | Time | Part amount |
| 010Y | 2015/5/1 | 123112 |
| 010Y | 2015/5/2 | 120644 |
| 010Y | 2015/5/3 | 137104 |
| 010Y | 2015/5/4 | 384792 |
| 010Y | 2015/5/5 | 397098 |
| 010Y | 2015/5/6 | 398597 |
| 010Y | 2015/5/7 | 402068 |
| 010Y | 2015/5/8 | 400924 |
| 。。。。。 | 。。。。。。 | 。。。。。 |
| 010Y | 2017/7/28 | 139777 |
Step 13, it is based on historical data, establishes the prediction model for predicting timing.
Historical data includes to meet the part amount of setting rule in logistic industry in the first setting duration.
Wherein, the first setting duration can be 1 year, is also possible to one month, can specifically carry out according to actual needs really
It is fixed.Meet setting rule, refers to and carry out statistical disposition for addressee amount or outbox amount.
Further, for continuing the elaboration in above-mentioned steps 12, which is 2015/05/01-
2017/07/28。
In above-mentioned steps 13, specific process flow is as shown in Fig. 2, as following:
Step 21, be based on historical data, respectively the growth trend of determining member amount, part amount mechanical periodicity and it is special when
Impact of the phase to part amount.
Wherein, in the technical solution that the embodiment of the present invention proposes, when establishing prediction model, Generalized Additive Models are based on
On the basis of Prophet model, each component is reset, specific as follows:
The growth trend g (t) of part amount can also be referred to as the long-term trend of part amount, pass through analyte stream industry part quantitative change
The characteristic of change obtains, which linearly increases；Slope and intercept are obtained by doing recurrence to all the points, to extract
The growth trend g (t) of part amount.
It is traditional to be achieved in that extracting long-term trend is by selecting minimum value and most for the growth trend g (t) of part amount
Big value line looks for slope, but this kind of mode and the temporal aspect for not meeting part amount in logistic industry, passes through analyte stream industry
The characteristic of part amount variation obtains, which linearly increases, therefore in the technical solution that the embodiment of the present invention proposes, leads to
It crosses and recurrence acquisition slope and intercept is done to all the points, pass through the fitting to linear increase curve come the growth trend g of determining member amount
(t)。
The mechanical periodicity s (t) of part amount can also be referred to as periodically.
In logistic industry, the polycyclic of part amount variation, in the technical solution that the embodiment of the present invention proposes, which is set
It is set to four kinds of periods, the i.e. period (T=365.25) in year, the period (T=30.4) of the moon, the period (T=7) in week, function spectrogram hair
Existing specific cycle (T=3.5).Further, the Fourier of year, the moon, week and specific cycle are respectively indicated using mathematic(al) manipulation
Series.
Impact h (t) of the particular time to part amount.
Particular time is the period that finger amount flies up or declines suddenly.It can also be referred to as festivals or holidays, but the section
Holiday has been different from festivals or holidays as defined in general law, also comprising red-letter day as defined in each electric business platform, the characteristics of those red-letter days
It is addressee amount and sends part amount that can all be multiplied.
Such as when the festivals or holidays (such as May Day, 11) of downward trend are overlapped with the date of uptrending in the period in week
When, remove the superposition of the cycle effect in its week.
Festivals or holidays effect indicates with indicative function, meets the holiday table of logistics business feature to going through according to design in advance
History is labeled on the time, obtains what component h (t) was made of 0 and 1, is 0 expression non-festivals or holidays on the time, on the time
Festivals or holidays are indicated for 1.
Step 22, add up the impact of growth trend, mechanical periodicity and particular time to part amount.
Step 23, prediction model of the accumulation result obtained as part amount.
Specifically, it can be indicated by following formula 1:
Y (t)=g (t)+s (t)+h (t)+l (t) formula 1
Wherein, y (t) is predicted value, and it be the mechanical periodicity of part amount, l (t) is special that g (t), which is the growth trend of part amount, s (t),
Different impact of the period to part amount.
Optionally, after above-mentioned steps 24, as shown in figure 3, can also include:
Determine the part amount and the part amount in historical data that lag item, lag item can also be referred to as hysteresis effect, be prediction
Distance, distance nearlyr weight it is bigger.
Hysteresis effect l (t) can also be referred to as lag item.The relationship of the part amount of time span of forecast and history part amount before, distance
Nearlyr weight is bigger.Think that history part amount has an impact to current part amount, and in time apart from current closer influence
It is bigger.By autocorrelation analysis, lag issue is obtained, autoregression model is fitted.
In view of lagging small battalion, it is possible to further be based on historical data, growth trend, the part amount of difference determining member amount
The impact to part amount of mechanical periodicity, particular time, and lag item, add up growth trend, mechanical periodicity, particular time are to group
The impact of part amount and lag item；Obtained accumulation result is as the prediction model for sending part amount.
Y (t)=g (t)+s (t)+h (t)+l (t)+εtFormula 2
Wherein, y (t) is predicted value, and it be the mechanical periodicity of part amount, l (t) is special that g (t), which is the growth trend of part amount, s (t),
Different impact of the period to part amount, εtIt is lag item.
Specifically, in above-mentioned process flow, impact of the particular time to part amount is determined, comprising:
Determine the time of part amount variation abnormality in historical data, wherein variation abnormality is that is increased beyond in finger amount curve
The situation of the second setting value was looked into one setting value or decline；The time of variation abnormality is divided into different time windows；According to
Time window determines impact of the particular time to part amount.
Specifically, it is described in detail with an example:
To consider festivals or holidays exogenous shock effect in prediction, festivals or holidays can be modeled by modifying festivals or holidays table.
Since the influence direction that festivals or holidays change part amount is different (upward or downward), and the festivals or holidays of each country are with their own characteristics, because
This design is applicable in loglstics enterprise and combines the festivals or holidays table of various countries' characteristic particularly important.Specifically:
The period for passing through analysis of history part amount variation abnormality first, to mark Chinese red-letter day and shopping section (such as 6.18, double ten
One)；
Secondly consider festivals or holidays on part amount influences it is preposition with delay effect, all red-letter days are finely tuned, section is divided into
Before, section neutralize section after.
The festivals or holidays for specifically including have: New Year's Day, Clear and Bright, the Dragon Boat Festival, May Day, shopping section 6.18, mid-autumn, National Day, double 11, double
12, the Spring Festival.Festivals or holidays effect is indicated with indicative function, according to the holiday table for meeting logistics business feature designed herein
(table 2) obtains what festivals or holidays component h (t) was made of 0 and 1, indicates non-festivals or holidays on the time for 0, is 1 table on the time
Show festivals or holidays.) and the table coverage is set to festivals or holidays effect, so that the time of festivals or holidays is extended to a section
[lower_window, upper_window], such as effect before the Mid-autumn Festival, then being also added on 7 days time before the Mid-autumn Festival
Into the holiday effect in the Mid-autumn Festival, the component of the period is set to 1.By taking the Mid-autumn Festival as an example, referring specifically to following table 2.
Table 2
| ds | holiday | lower_window | upper_window |
| 2015/6/17 | BeforeMA | 0 | 7 |
| 2016/6/6 | BeforeMA | 0 | 7 |
| 2017/5/25 | BeforeMA | 0 | 7 |
| 2015/9/26 | MidAutumn | 0 | 2 |
| 2016/9/15 | MidAutumn | 0 | 2 |
| 2017/10/4 | MidAutumn | 0 | 2 |
| 2015/9/29 | AfterMA | 0 | 1 |
| 2016/9/18 | AfterMA | 0 | 1 |
| 2017/10/7 | AfterMA | 0 | 1 |
Step 14, it is based on prediction model, the part amount in logistic industry is predicted.
Specifically, prediction model is established according to historical data, wherein include in prediction model it is each when order components, such as on
Long-term trend, period, festivals or holidays and lag item, each component that text illustrates are added to obtain history part amount daily on historical time.
The parameter value of each component is obtained by doing training to historical data, can be obtained on the corresponding time multiplied by the time t of time span of forecast respectively
Part amount predicted value.
In technical solution set forth above of the embodiment of the present invention, by big data analysis technology, with going through in certain time length
History data are foundation, are analyzed those historical datas, and then determine prediction model, are carried out by prediction model to part amount
Prediction, avoids manual type to the statistics of part amount, for logistic industry, historical data is more, trained prediction model
Also more accurate, obtained result is also more accurate, can preferably how much be prejudged for part amount, so as to send part, receipts
Part carries out reasonable distribution, and saving is artificial, improves efficiency.
Correspondingly, the embodiment of the present invention also proposes a kind of time series data prediction meanss, as shown in figure 3, device includes:
Module 201 is obtained, for obtaining the historical data for meeting setting rule in logistic industry in the first setting duration；
Processing module 202 establishes the prediction model for predicting timing for being based on the historical data；And it is based on
The prediction model predicts the part amount in logistic industry.
The historical data includes to meet the part amount of setting rule in logistic industry in the first setting duration；The processing
Module is specifically used for being based on the historical data, respectively the growth trend of determining member amount, the mechanical periodicity of part amount and special
Impact of the period to part amount, wherein the particular time is the period that finger amount flies up or declines suddenly；Add up the increasing
The impact of long trend, mechanical periodicity and particular time to part amount；Prediction model of the obtained accumulation result as part amount.
The acquisition module is also used to determine that lag item, the lag item refer to the explained variable of phase in the past to current
Explained variable have an impact；The processing module is also used to based on the historical data, and the growth of determining member amount becomes respectively
The impact of gesture, the mechanical periodicity of part amount, particular time to part amount, and lag item, wherein the particular time is that finger amount is prominent
The period for so rising or declining suddenly；The growth trend, mechanical periodicity, particular time add up to the impact of sending part amount and stagnant
It is consequent；Obtained accumulation result is as the prediction model for sending part amount.
The processing module, specifically for the time of part amount variation abnormality in the determination historical data, wherein the change
Changing abnormal is the first setting value to be increased beyond in finger amount curve or the situation of the second setting value was looked into decline；The variation is different
The normal time is divided into different time windows；According to the time window, impact of the particular time to part amount is determined.
Correspondingly, the present invention also proposes a kind of computer readable storage medium, is stored thereon with computer program, feature
The step of being, time series data prediction technique realized when the computer program is executed by processor.
Flow chart and block diagram in attached drawing are illustrated according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yiyong
The dedicated hardware based system of defined functions or operations is executed to realize, or can be referred to specialized hardware and computer
The combination of order is realized.
Being described in the embodiment of the present application involved unit or module can be realized by way of software, can also be with
It is realized by way of hardware.Described unit or module also can be set in the processor.These units or module
Title does not constitute the restriction to the unit or module itself under certain conditions.
As on the other hand, present invention also provides a kind of computer readable storage medium, the computer-readable storage mediums
Matter can be computer readable storage medium included in device described in above-described embodiment；It is also possible to individualism, not
The computer readable storage medium being fitted into equipment.Computer-readable recording medium storage has one or more than one journey
Sequence, described program are used to execute the formula input method for being described in the application by one or more than one processor.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from the inventive concept, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of time series data prediction technique, which is characterized in that the described method includes:
Obtain the historical data for meeting setting rule in logistic industry in the first setting duration；
Based on the historical data, the prediction model for predicting timing is established；
Based on the prediction model, the part amount in logistic industry is predicted.
2. the method as described in claim 1, which is characterized in that the historical data include logistic industry in first setting when
Meet the part amount of setting rule in length；
Based on the historical data, the prediction model for predicting timing is established, comprising:
Based on the historical data, the growth trend of determining member amount, the mechanical periodicity of part amount and particular time are to part amount respectively
Impact, wherein the particular time is the period that finger amount flies up or declines suddenly；
Add up the impact of the growth trend, mechanical periodicity and particular time to part amount；
Prediction model of the obtained accumulation result as part amount.
3. method according to claim 2, which is characterized in that further include:
Determine that lag item, the lag item refer to that the explained variable of phase in the past has an impact current explained variable；
Based on the historical data, the prediction model for predicting timing is established, comprising:
Based on the historical data, the growth trend of determining member amount, the mechanical periodicity of part amount, particular time rush part amount respectively
It hits, and lag item, wherein the particular time is the period that finger amount flies up or declines suddenly；
Impact and lag item of the growth trend, mechanical periodicity, particular time of adding up to part amount is sent；
Obtained accumulation result is as the prediction model for sending part amount.
4. method as claimed in claim 2 or claim 3, which is characterized in that determine impact of the particular time to part amount, comprising:
The time of part amount variation abnormality in the historical data is determined, wherein the variation abnormality is to rise to surpass in finger amount curve
It crosses the first setting value or the situation of the second setting value was looked into decline；
The time of the variation abnormality is divided into different time windows；
According to the time window, impact of the particular time to part amount is determined.
5. the method as described in claim 1, which is characterized in that the part amount includes the addressee amount of logistic industry；Or
Logistic industry sends part amount.
6. a kind of time series data prediction meanss, which is characterized in that described device includes:
Module is obtained, for obtaining the historical data for meeting setting rule in logistic industry in the first setting duration；
Processing module establishes the prediction model for predicting timing for being based on the historical data；And it is based on the prediction
Model predicts the part amount in logistic industry.
7. device as claimed in claim 6, which is characterized in that the historical data include logistic industry in first setting when
Meet the part amount of setting rule in length；
The processing module is specifically used for being based on the historical data, respectively the period change of the growth trend of determining member amount, part amount
The impact of change and particular time to part amount, wherein the particular time is the period that finger amount flies up or declines suddenly；
Add up the impact of the growth trend, mechanical periodicity and particular time to part amount；Obtained accumulation result is pre- as part amount
Survey model.
8. device as claimed in claim 7, which is characterized in that the acquisition module is also used to determine lag item, the lag
Item refers to that the explained variable of phase in the past has an impact current explained variable；
The processing module is also used to based on the historical data, respectively the period change of the growth trend of determining member amount, part amount
Change, impact of the particular time to part amount, and lag item, wherein the particular time is that finger amount flies up or declines suddenly
Period；Impact and lag item of the growth trend, mechanical periodicity, particular time of adding up to part amount is sent；What is obtained is cumulative
As a result as the prediction model for sending part amount.
9. device as claimed in claim 7 or 8, which is characterized in that the processing module is specifically used for determining the history number
According to the time of middle part amount variation abnormality, wherein the variation abnormality is to be increased beyond the first setting value or decline in finger amount curve
Looked into the situation of the second setting value；The time of the variation abnormality is divided into different time windows；According to the time window
Mouthful, determine impact of the particular time to part amount.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of time series data prediction technique as described in any one of claims 1 to 5 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711456483.6A CN109978208A (en) | 2017-12-28 | 2017-12-28 | A kind of time series data prediction technique, device and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711456483.6A CN109978208A (en) | 2017-12-28 | 2017-12-28 | A kind of time series data prediction technique, device and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109978208A true CN109978208A (en) | 2019-07-05 |
Family
ID=67074318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711456483.6A Pending CN109978208A (en) | 2017-12-28 | 2017-12-28 | A kind of time series data prediction technique, device and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109978208A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110443425A (en) * | 2019-08-09 | 2019-11-12 | 长江慧控科技(武汉)有限公司 | Intelligent railway station electric energy energy consumption prediction technique based on Prophet |
| CN112418898A (en) * | 2019-08-21 | 2021-02-26 | 北京京东乾石科技有限公司 | Article demand data analysis method and device based on multi-time window fusion |
| CN112862137A (en) * | 2019-11-27 | 2021-05-28 | 顺丰科技有限公司 | Method and device for predicting quantity, computer equipment and computer readable storage medium |
| CN112907267A (en) * | 2019-12-03 | 2021-06-04 | 顺丰科技有限公司 | Method and device for predicting cargo quantity, computer equipment and storage medium |
| CN112906930A (en) * | 2019-12-04 | 2021-06-04 | 顺丰科技有限公司 | Site cargo quantity prediction method, device, equipment and storage medium |
| CN112990526A (en) * | 2019-12-16 | 2021-06-18 | 顺丰科技有限公司 | Method and device for predicting logistics arrival quantity and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955625A (en) * | 2014-05-21 | 2014-07-30 | 河海大学 | Determination method of nonlinear Almon distributed lag |
| CN109657831A (en) * | 2017-10-11 | 2019-04-19 | 顺丰科技有限公司 | A kind of Traffic prediction method, apparatus, equipment, storage medium |
-
2017
- 2017-12-28 CN CN201711456483.6A patent/CN109978208A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955625A (en) * | 2014-05-21 | 2014-07-30 | 河海大学 | Determination method of nonlinear Almon distributed lag |
| CN109657831A (en) * | 2017-10-11 | 2019-04-19 | 顺丰科技有限公司 | A kind of Traffic prediction method, apparatus, equipment, storage medium |
Non-Patent Citations (1)
| Title |
|---|
| SHANGFR: "中国快递包裹总量：基于SARIMA模型的预测", 《HTTPS://WWW.CNBLOGS.COM/SHANGFR/P/4919071.HTML》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110443425A (en) * | 2019-08-09 | 2019-11-12 | 长江慧控科技(武汉)有限公司 | Intelligent railway station electric energy energy consumption prediction technique based on Prophet |
| CN112418898A (en) * | 2019-08-21 | 2021-02-26 | 北京京东乾石科技有限公司 | Article demand data analysis method and device based on multi-time window fusion |
| CN112862137A (en) * | 2019-11-27 | 2021-05-28 | 顺丰科技有限公司 | Method and device for predicting quantity, computer equipment and computer readable storage medium |
| CN112907267A (en) * | 2019-12-03 | 2021-06-04 | 顺丰科技有限公司 | Method and device for predicting cargo quantity, computer equipment and storage medium |
| CN112906930A (en) * | 2019-12-04 | 2021-06-04 | 顺丰科技有限公司 | Site cargo quantity prediction method, device, equipment and storage medium |
| CN112990526A (en) * | 2019-12-16 | 2021-06-18 | 顺丰科技有限公司 | Method and device for predicting logistics arrival quantity and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109978208A (en) | A kind of time series data prediction technique, device and computer storage medium | |
| Sellner et al. | Air accessibility and growth–The economic effects of a capacity expansion at Vienna International Airport | |
| CN108846493B (en) | Method for predicting air ticket booking number based on improved incremental model | |
| CN110019401B (en) | Method, device, equipment and storage medium for predicting part quantity | |
| JP2015125775A (en) | System and method for multi-task learning system for prediction of demand on system | |
| Josélyne et al. | Partitioning microservices: A domain engineering approach | |
| CN105512773A (en) | Passenger travel destination prediction method and device | |
| CN105869017A (en) | Method and system for predicting ticket prices | |
| CN109308551A (en) | A kind of certain region electric energy substitution Potentials method based on GM gray prediction | |
| CN111161566B (en) | Flight delay early warning method, system, electronic equipment and medium | |
| Albliwi et al. | Implementation of a lean six sigma approach in the manufacturing Sector: a systematic literature review | |
| CN109993645A (en) | Financial forecast method, system, computer system and computer readable storage medium | |
| CN104834730A (en) | Data analysis system and method | |
| Tyrin et al. | Multi-agent system “Smart Factory” for real-time workshop management: Results of design & implementation for Izhevsk Axion-Holding Factory | |
| Bianchi et al. | Load forecasting in district heating networks: Model comparison on a real-world case study | |
| CN106202106A (en) | Method and system is recommended in a kind of efficient data analysis | |
| CN110309399A (en) | Scene information method for pushing and device based on date classification | |
| Okeke et al. | Foreign direct investment and economic growth in Nigeria: An empirical evidence | |
| Pradhan et al. | Time series moving average, smoothing analysis, forecasting analysis and evaluation for natural gas consumption | |
| Chèze 2 3 4 et al. | Air traffic energy efficiency differs from place to place: New results from a macro-level approach 1 | |
| Gorican et al. | Knowledge transfer ontology of photovoltaic electricity production forecasting | |
| Caceres-Hernandez et al. | A proposal for measuring and comparing seasonal variations in hourly economic time series | |
| Bianchi et al. | XM_HeatForecast: Heating Load Forecasting in Smart District Heating Networks | |
| Sison et al. | Survey of Machine Learning and Deep Learning Techniques for Travel Demand Forecasting | |
| Poerwanto et al. | The Ergonomic Analysis of the Airline Passengers Message Service to Improve the Flight Safety |
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: 20190705 |