CN107506869A - A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network - Google Patents
A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network Download PDFInfo
- Publication number
- CN107506869A CN107506869A CN201710792903.1A CN201710792903A CN107506869A CN 107506869 A CN107506869 A CN 107506869A CN 201710792903 A CN201710792903 A CN 201710792903A CN 107506869 A CN107506869 A CN 107506869A
- Authority
- CN
- China
- Prior art keywords
- data
- date
- day
- booking
- course line
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 18
- 230000000306 recurrent effect Effects 0.000 title claims abstract description 16
- 238000000034 method Methods 0.000 title claims abstract description 13
- 230000007935 neutral effect Effects 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 9
- 230000003203 everyday effect Effects 0.000 claims description 8
- 241001269238 Data Species 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 210000004218 nerve net Anatomy 0.000 claims 1
- 230000010006 flight Effects 0.000 abstract description 6
- 230000005540 biological transmission Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010297 mechanical methods and process Methods 0.000 description 1
- 230000005226 mechanical processes and functions Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000011160 research 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G06Q50/40—
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to number Forecasting Methodology of making a reservation a kind of air ticket day based on Recognition with Recurrent Neural Network.Air ticket day booking number Forecasting Methodology of the present invention based on Recognition with Recurrent Neural Network, using Recognition with Recurrent Neural Network modeling aviation booking data sequence information, applied to aviation booking data prediction;Compared with prior art, total ticket booking number when all flights in certain course line finally take off can not only be predicted, the air ticket quantity that each advance ticket date sells before flight takeoff can also be predicted, and then the ticket price on each presell date is instructed to formulate, to reach the purpose for making airline's maximum revenue, there is important practical application meaning.
Description
Technical field
The present invention relates to number Forecasting Methodology of making a reservation a kind of air ticket day based on Recognition with Recurrent Neural Network, belong to Civil Aviation IT
Technical field.
Background technology
The target of yield management be by commodity in the suitable time with suitable price sales to suitable customer, to reach
The final goal of maximum revenue.As the key technology of airline revenue mangement, course line passenger flow forecast is that airline implements
The basis of the operations such as Dynamic Pricing, the control of seat storage, is one of important process of airline's development plan.Accurate passenger flow
Amount prediction can provide irreplaceable decision support for the development of airline.
The current existing research on course line passenger flow forecast, it is to utilize mathematics according to flight history booking data mostly
Final volume of the flow of passengers when model is to predict the flight takeoff.In the actual mechanical process of airline, it is final to predict flight
Although total ticketing number for instruct vehicle supervision department ahead of time planned and managed and optimize aeronautical resources configure have
Important function, but for instructing having little significance for the air ticket Dynamic Pricing of the flight presell interim every day.And if can be accurate
The air ticket quantity that the flight in the course line is sold interim every day in presell is really predicted, then to instructing every day flight presell phase
Air ticket pricing it is significant.Airline formulates air ticket Dynamic Pricing plan based on the daily booking number predicted
Slightly, to reach the purpose of maximum gain.
Artificial neural network is the operational model being interconnected to constitute by a large amount of neurons, in pattern-recognition, intelligent machine
Device people, the field such as automatically control and successfully solve the insoluble practical problem of many modern computers.Recognition with Recurrent Neural Network
(RNN) it is a kind of artificial neural network of particular design, the difference of itself and conventional feed forward neutral net is, its hidden layer
Directed circulation is introduced, the information transmission of last moment to current time can be handled front and rear pass between those inputs
The problem of connection.
The content of the invention
In view of the shortcomings of the prior art, the present invention provides a kind of air ticket day booking number prediction side based on Recognition with Recurrent Neural Network
Method.
The technical scheme is that:
A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network, including step are as follows:
1) initial data is pre-processed;Statistics belongs to the flight in same course line, total freight space number in every course line and every
Total booking number in bar course line;
Wherein, every course line has n+1 bars to make a reservation data, and n+1 bars booking data correspond to before flight takeoff n days to flight
Take off the same day every day course line booking number;Flight therein is all flights that a certain course line includes.
The presell phase of usual flight be from n days before the flight takeoff to the flight takeoff on the day of, therefore for each boat
Class, we have n+1 bars booking data;What we to be predicted is the booking data in certain course line, rather than some specific flight
Booking data, it is therefore desirable to data are pre-processed, the data message of all flights in the course line in every day is summed,
Obtain the total data in the course line.
2) it is first, n+1 groups according to the date difference between advance ticket date and date by total data, numbering i=0
~n;N+1 groups data correspond in the different ticketing dates total freight space number in every course line and total booking number in every course line respectively;
To each group of data, the data of continuous seven dates are taken, wherein the data of the first six date are as god
Input through network, the output of the data of the 7th date as neutral net;The form of the input of neutral net is 6 rows
The matrix of 4 row, every data line of matrix represent the historical data of date;4 column datas of matrix refer respectively to the group
Whether the numbering i of data, the date are the ticketing sum of i days, the cabin of the 7th day to be predicted before weekend, the date
Position sum;Wherein, group # i and the freight space sum of the 7th day to be predicted are that 6 rows share;The form of the output of neutral net
For for a scalar value;The scalar value of neutral net output represents the course line of the 7th day to be predicted and worked as when taking off i days a few days ago
It ticketing number.It is identical that so-called 6 row shares i.e. 6 row data.
3) above operation is carried out successively to whole n+1 groups data, obtains whole training data samples;
4) test sample is chosen, inputs in the model trained, obtains prediction result.
According to currently preferred, the n=17.
Beneficial effects of the present invention are:
1. the air ticket day booking number Forecasting Methodology of the present invention based on Recognition with Recurrent Neural Network, is built using Recognition with Recurrent Neural Network
Mould aviation booking data sequence information, applied to aviation booking data prediction;Compared with prior art, certain boat can not only be predicted
Total ticket booking number when all flights of line finally take off, moreover it is possible to the air ticket quantity that each advance ticket date sells before flight takeoff
It is predicted, and then instructs the ticket price on each presell date to formulate, to reach the purpose for making airline's maximum revenue,
With important practical application meaning;
2.RNN introduces directed circulation in hidden layer, can locate the information transmission of last moment to current time
Manage those input between forward-backward correlation the problem of this property, make its be particularly suitable for handle sequence information;And certain course line is gone through
The certain sequential relationship of history ticket booking data fit, we can learn the sequential of course line ticket booking data using Recognition with Recurrent Neural Network
Rule, the purpose that current ticket booking number is predicted according to historical data is realized well.
Brief description of the drawings
Fig. 1 is number Forecasting Methodology flow chart of making a reservation the air ticket day of the present invention based on Recognition with Recurrent Neural Network;
Fig. 2 is the illustraton of model of neutral net of the present invention.
Embodiment
With reference to embodiment and Figure of description, the present invention will be further described, but not limited to this.
Embodiment 1
As shown in Figure 1-2.
A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network, including step are as follows:
1) initial data is pre-processed;Statistics belongs to the flight in same course line, total freight space number in every course line and every
Total booking number in bar course line;
Wherein, every course line has n+1 bars to make a reservation data, and n+1 bars booking data correspond to before flight takeoff n days to flight
Take off the same day every day course line booking number;Flight therein is all flights that a certain course line includes.
The presell phase of usual flight be from n days before the flight takeoff to the flight takeoff on the day of, therefore for each boat
Class, we have n+1 bars booking data;What we to be predicted is the booking data in certain course line, rather than some specific flight
Booking data, it is therefore desirable to data are pre-processed, the data message of all flights in the course line in every day is summed,
Obtain the total data in the course line.
2) it is first, n+1 groups according to the date difference between advance ticket date and date by total data, numbering i=0
~n;N+1 groups data correspond in the different ticketing dates total freight space number in every course line and total booking number in every course line respectively;
To each group of data, the data of continuous seven dates are taken, wherein the data of the first six date are as god
Input through network, the output of the data of the 7th date as neutral net;The form of the input of neutral net is 6 rows
The matrix of 4 row, every data line of matrix represent the historical data of date;4 column datas of matrix refer respectively to the group
Whether the numbering i of data, the date are the ticketing sum of i days, the cabin of the 7th day to be predicted before weekend, the date
Position sum;Wherein, group # i and the freight space sum of the 7th day to be predicted are that 6 rows share;The form of the output of neutral net
For for a scalar value;The scalar value of neutral net output represents the course line of the 7th day to be predicted and worked as when taking off i days a few days ago
It ticketing number.It is identical that so-called 6 row shares i.e. 6 row data.
3) above operation is carried out successively to whole n+1 groups data, obtains whole training data samples;
4) test sample is chosen, inputs in the model trained, obtains prediction result.
Wherein, n=17.
Claims (2)
1. a kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network, it is characterised in that as follows including step:
1) initial data is pre-processed;Statistics belongs to the flight in same course line, total freight space number in every course line and every boat
Total booking number of line;
Wherein, every course line has n+1 bars to make a reservation data, and n+1 bars booking data correspond to before flight takeoff n days to flight takeoff
Every day on the same day course line booking number;
2) it is first, n+1 groups according to the date difference between advance ticket date and date by total data, numbering i=0~n;
N+1 groups data correspond in the different ticketing dates total freight space number in every course line and total booking number in every course line respectively;
To each group of data, the data of continuous seven dates are taken, wherein the data of the first six date are as nerve net
The input of network, the output of the data of the 7th date as neutral net;The form of the input of neutral net is that 6 rows 4 arrange
Matrix, every data line of matrix represents the historical data of date;4 column datas of matrix refer respectively to this group of data
Numbering i, the ticketing the sum whether date is i days before weekend, the date, the freight space of the 7th day to be predicted it is total
Number;Wherein, group # i and the freight space sum of the 7th day to be predicted are that 6 rows share;The form of the output of neutral net be for
One scalar value;
3) above operation is carried out successively to whole n+1 groups data, obtains whole training data samples;
4) test sample is chosen, inputs in the model trained, obtains prediction result.
2. the air ticket day booking number Forecasting Methodology according to claim 1 based on Recognition with Recurrent Neural Network, it is characterised in that institute
State n=17.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710792903.1A CN107506869A (en) | 2017-09-05 | 2017-09-05 | A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710792903.1A CN107506869A (en) | 2017-09-05 | 2017-09-05 | A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107506869A true CN107506869A (en) | 2017-12-22 |
Family
ID=60695711
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710792903.1A Pending CN107506869A (en) | 2017-09-05 | 2017-09-05 | A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107506869A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109858671A (en) * | 2018-12-26 | 2019-06-07 | 携程旅游网络技术(上海)有限公司 | Predict the method and system of the visiting rate of target pattern |
CN110516873A (en) * | 2019-08-28 | 2019-11-29 | 贵州优策网络科技有限公司 | A kind of airline's Slot Allocation optimization method |
CN111192090A (en) * | 2019-12-31 | 2020-05-22 | 广州优策科技有限公司 | Seat allocation method and device for flight, storage medium and electronic equipment |
CN111899059A (en) * | 2020-08-12 | 2020-11-06 | 科技谷(厦门)信息技术有限公司 | Navigation driver revenue management dynamic pricing method based on block chain |
CN112308618A (en) * | 2020-11-02 | 2021-02-02 | 沈阳民航东北凯亚有限公司 | Data processing method and device, electronic equipment and storage medium |
CN112396243A (en) * | 2020-11-30 | 2021-02-23 | 中国民航信息网络股份有限公司 | Flight booking value processing method and system based on addition model |
CN112907158A (en) * | 2021-05-10 | 2021-06-04 | 北京人人云图信息技术有限公司 | Flight passenger flow generation method and system based on game learning |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105023068A (en) * | 2015-08-06 | 2015-11-04 | 北京航空航天大学 | Rule mining based flight arrival and departure cooperative scheduling method |
-
2017
- 2017-09-05 CN CN201710792903.1A patent/CN107506869A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105023068A (en) * | 2015-08-06 | 2015-11-04 | 北京航空航天大学 | Rule mining based flight arrival and departure cooperative scheduling method |
Non-Patent Citations (2)
Title |
---|
ALICE熹爱学习: "用LSTM做时间序列预测的一个小例子", 《HTTPS://BLOG.CSDN.NET/ALICEYANGXI1987/ARTICLE/DETAILS/73420583》 * |
张沂华: "基于灰色拓扑的航班订座需求预测方法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109858671A (en) * | 2018-12-26 | 2019-06-07 | 携程旅游网络技术(上海)有限公司 | Predict the method and system of the visiting rate of target pattern |
CN109858671B (en) * | 2018-12-26 | 2021-06-18 | 携程旅游网络技术(上海)有限公司 | Method and system for predicting passenger seat rate of target airline |
CN110516873A (en) * | 2019-08-28 | 2019-11-29 | 贵州优策网络科技有限公司 | A kind of airline's Slot Allocation optimization method |
CN111192090A (en) * | 2019-12-31 | 2020-05-22 | 广州优策科技有限公司 | Seat allocation method and device for flight, storage medium and electronic equipment |
CN111899059A (en) * | 2020-08-12 | 2020-11-06 | 科技谷(厦门)信息技术有限公司 | Navigation driver revenue management dynamic pricing method based on block chain |
CN112308618A (en) * | 2020-11-02 | 2021-02-02 | 沈阳民航东北凯亚有限公司 | Data processing method and device, electronic equipment and storage medium |
CN112396243A (en) * | 2020-11-30 | 2021-02-23 | 中国民航信息网络股份有限公司 | Flight booking value processing method and system based on addition model |
CN112907158A (en) * | 2021-05-10 | 2021-06-04 | 北京人人云图信息技术有限公司 | Flight passenger flow generation method and system based on game learning |
CN112907158B (en) * | 2021-05-10 | 2021-07-23 | 北京人人云图信息技术有限公司 | Flight passenger flow generation method and system based on game learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107506869A (en) | A kind of air ticket day booking number Forecasting Methodology based on Recognition with Recurrent Neural Network | |
CN107086935B (en) | People flow distribution prediction method based on WIFI AP | |
CN110458336B (en) | Online appointment vehicle supply and demand prediction method based on deep learning | |
Chau | Application of a PSO-based neural network in analysis of outcomes of construction claims | |
CN104751681B (en) | Statistical learning model based gate position allocation method | |
Liu et al. | Scheduling optimisation of multi-type special vehicles in an airport | |
CN109767128A (en) | imaging satellite autonomous task planning method based on machine learning | |
CN111754025A (en) | Public transport short-time passenger flow prediction method based on CNN + GRU | |
Zhao et al. | Evaluation of the airport ferry vehicle scheduling based on network maximum flow model | |
CN112330983A (en) | Integrated intelligent recovery method for abnormal flight | |
Cai et al. | A bi-objective constrained robust gate assignment problem: Formulation, instances and algorithm | |
CN109583659A (en) | User's operation behavior prediction method and system based on deep learning | |
Liu et al. | Stochastic airline fleet assignment with risk aversion | |
CN112132366A (en) | Prediction system for flight clearance rate | |
She et al. | A robust strategy to address the airport gate assignment problem considering operators’ preferences | |
Graham et al. | Estimating the productivity of cyclic construction operations using case-based reasoning | |
CN115222251A (en) | Network taxi appointment scheduling method based on hybrid layered reinforcement learning | |
CN111340536A (en) | Model training method, passenger seat rate progress prediction method, system, device and medium | |
CN113269402B (en) | Flight space control method and device and computer equipment | |
CN114330842A (en) | Airport ferry vehicle scheduling method based on two-stage optimization | |
CN110659773A (en) | Flight delay prediction method based on deep learning | |
Chen et al. | A simple process simulation model for strategic planning on the airside of an airport: a case study | |
CN112101647A (en) | Multi-objective optimization method for flight test efficiency cost of civil transport aircraft | |
Gonzalez | Robust evolutionary methods for multi-objective and multdisciplinary design optimisation in aeronautics | |
Mumayiz | A methodology for planning and operations management of airport passenger terminals: a capacity/level of service approach |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171222 |
|
RJ01 | Rejection of invention patent application after publication |