Flight passenger flow rate prediction method based on multi-granularity time attention mechanism

A technology of attention and passenger load factor, applied in forecasting, neural learning methods, instruments, etc., can solve problems that cannot be considered at the same time

Inactive Publication Date: 2019-09-27
BEIJING JIAOTONG UNIV
View PDF2 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] At present, the existing flight passenger load factor prediction methods only focus on the recent changes in the daily passenger load factor of the flight, and cannot take into account the impact of many other factors on the problem, such as the passenger load factor of other flights near the departure time of the flight route. , the attributes of the flight itself and external factors, etc.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Flight passenger flow rate prediction method based on multi-granularity time attention mechanism
  • Flight passenger flow rate prediction method based on multi-granularity time attention mechanism
  • Flight passenger flow rate prediction method based on multi-granularity time attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] The embodiment of the present invention proposes a recurrent neural network model (MTA-RNN) based on a multi-granularity temporal attention mechanism, figure 1 An architecture diagram of the above-mentioned MTA-RNN model provided for the embodiment of the present invention, the model constructs a multi-level attention mechanism to obtain the time series correlation of the flight passenger load factor at different time granularities, the model includes: combined with the attention mechanism at the time of departure The encoder and the decoder combined with the off-day attention mechanism.

[0051] The encoder that combines the attention mechanism at the moment of take-off includes a hierarchical structure of the first layer of long short-term memory network LSTM unit and the second layer of LSTM unit, the first layer of LSTM unit is figure 1 The LSTM unit below the encoder in , and the second layer LSTM unit is figure 1 LSTM unit above the encoder in . The historical p...

Embodiment 2

[0092] figure 2 This is a comparison chart of economy class occupancy factors of flights with different departure times on the Beijing-Shanghai route in March 2010. The horizontal axis is the flight departure date, and the vertical axis is the economy class occupancy factors of flights. Different curves represent changes in economy class occupancy factors of flights at different departure times curve. From figure 2 It can be seen that the passenger load factor of the flight has the following characteristics:

[0093] 1. Relevance of take-off time. Given a certain route, the passenger load factors of flights with different departure times on a certain departure day affect each other, and the passenger load factors of flights with closer departure times are closer. Depend on figure 2 It can be seen that the daily economy class passenger load factor and the overall passenger load factor curve of the 8:30AM flight and the 9:00AM flight with similar departure times are close...

Embodiment approach

[0098] Taking the real historical flight passenger load factor data set provided by TravelSky as an example, the flight data of the Beijing-Shanghai route with more than 584 (730*89%) departure days within two years is selected as the experimental data set. If the time is set The window length D=28, and the number of forecast days τ=7, then the statistical information is shown in Table 1.

[0099] Table 1 Dataset Statistics

[0100]

[0101] image 3 A processing flowchart of a flight load factor prediction method based on multi-granularity time attention provided by an embodiment of the present invention includes the following processing steps:

[0102] Step S1, constructing experimental samples based on the experimental data set, and dividing the training set, verification set and test set;

[0103]The present invention divides the data set into non-overlapping training, verification and test sets in chronological order, and the division ratio is 8:1:1, that is, 27 flig...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a flight passenger flow rate prediction method based on a multi-granularity time attention mechanism. The method comprises the following steps of: constructing a recurrent neural network model based on a multi-granularity time attention mechanism; taking the historical flight passenger rate time sequences of all take-off moments on the air route as input sequences of the encoders, enabling the encoders to encode the input sequences, enabling the decoders to decode the coded information output by the encoders, and obtaining a flight passenger rate time sequence of the target flight. According to the method, the time sequence dependence of flight passenger rates at different take-off moments and the influence of the flight passenger rates at other take-off moments on the target flight passenger rate in the air route where the target flight is located are captured through the take-off moment attention mechanism, and meanwhile the trend and periodicity of the passenger rate sequence of the target flight are captured through the take-off day attention mechanism; and the influence of external factors such as flight attributes and holidays and festivals is considered, so that the model has a good effect on flight passenger flow rate prediction.

Description

technical field [0001] The invention relates to the technical field of flight passenger load factor prediction, in particular to a flight passenger load factor prediction method based on a multi-granularity time attention mechanism. Background technique [0002] With the rapid growth of civil aviation passengers, the demand forecast of air passenger transportation is more and more concerned by airlines, ticket agents, aircraft manufacturers and other civil aviation-related enterprises. Air passenger traffic demand forecasting includes airline passenger traffic demand forecasting, airport passenger traffic demand forecasting, airline market share forecasting, and flight passenger traffic demand forecasting. Airline passenger traffic demand forecasting is a finer-grained demand forecasting, and it is the basis of airline seat optimization control and pricing strategy. [0003] For the air passenger transportation market, the passenger load factor of a flight is an important i...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04G06Q10/06G06Q50/30G06N3/04G06N3/08
CPCG06Q10/04G06Q10/06315G06Q50/30G06N3/08G06N3/044G06N3/045
Inventor 武志昊林友芳韩升万怀宇王晶董兴业张硕
Owner BEIJING JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap