CN106779245B - Event-based civil aviation demand prediction method and device - Google Patents
Event-based civil aviation demand prediction method and device Download PDFInfo
- Publication number
- CN106779245B CN106779245B CN201710004030.3A CN201710004030A CN106779245B CN 106779245 B CN106779245 B CN 106779245B CN 201710004030 A CN201710004030 A CN 201710004030A CN 106779245 B CN106779245 B CN 106779245B
- Authority
- CN
- China
- Prior art keywords
- civil aviation
- date
- current flight
- demand
- curve
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000002159 abnormal effect Effects 0.000 claims description 138
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012935 Averaging Methods 0.000 claims description 10
- 238000003064 k means clustering Methods 0.000 claims description 9
- 230000003442 weekly effect Effects 0.000 claims description 4
- 238000013277 forecasting method Methods 0.000 claims description 2
- 230000005856 abnormality Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 3
- 230000002547 anomalous effect Effects 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000010006 flight Effects 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
Images
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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a civil aviation demand prediction method and device based on events. Wherein, the method comprises the following steps: determining the prediction type of the current flight on the take-off date according to the take-off date of the current flight, and the influence factors of the exogenous events to be generated on the current date on the civil aviation demand of the current flight; and according to the prediction type of the current flight on the take-off date, predicting the day-level civil aviation demand prediction index of the current flight on the take-off date. According to the method and the device, the problem of inaccurate civil aviation demand prediction is solved, the influence of the external source event on the civil aviation demand is added, and the accuracy of the civil aviation demand prediction is improved.
Description
Technical Field
The invention relates to the field of civil aviation data processing, in particular to a method and a device for forecasting civil aviation demands based on events.
Background
In the field of civil aviation, the problem that always puzzles the airline revenue manager is: on the premise of keeping the maximum income, the reasonable arrangement of the bin positions is provided to meet the diversified and disorderly travel demands of the aviation passengers. With the rapid development of the civil aviation industry, the accurate prediction of the change of market demand is a core appeal of an airline revenue management department, wherein the prediction of passenger flight demand based on O & D (origin & arrival, also abbreviated as OD) is of great importance to airline revenue managers. Although airlines can easily obtain historical passenger volume data of each airline at present, no mature analysis and prediction model can reflect the future demand change of passengers for flights so as to provide decision support for revenue managers. Therefore, how to effectively predict the future change trend of the flight demand of the passengers in time and provide a decision basis for the revenue management of the airlines is a new stage requirement for civil aviation information service.
The current technology can mainly realize the inquiry function of the historical passenger volume of each flight based on O & D, the current passenger demand is reflected by historical data, the current demand trend can only be reflected in real time, and the effective prediction of the future demand change trend cannot be provided, so that the response time given to a revenue manager by the statistical model is very short. Even if a certain prediction function is provided, only the influence of historical data is considered, and the influence of objective events on demand change is ignored. In view of the current situation, domestic airlines mainly use traditional market demand measurement methods for medium and long-term sales management (generally referring to flights in two weeks or later), and short-term sales management (generally within two weeks) mainly depends on human experience, lacks prediction of variation trend of passenger flight demand, and provides effective data support for revenue managers.
Disclosure of Invention
The invention provides a civil aviation demand prediction method and device based on events, which at least solve the problem of inaccurate prediction of civil aviation demands in the related technology.
According to one aspect of the invention, an event-based civil aviation demand forecasting method is provided, which comprises the following steps: determining the prediction type of the current flight on the take-off date according to the take-off date of the current flight, and influence factors of exogenous events to be generated on the current date on the civil aviation demand of the current flight; and predicting the day-level civil aviation demand prediction index of the current flight on the take-off date according to the prediction type of the current flight on the take-off date.
Optionally, before determining the predicted type of the current flight on the takeoff date according to whether there is an abnormality in the takeoff date of the current flight and whether there is an event affecting civil aviation requirements on the takeoff date, the method further includes: determining whether the take-off date of the current flight is abnormal or not according to a week civil aviation demand standard curve of the airline where the current flight is located and a civil aviation demand abnormal curve of the airline; and acquiring an external source event which will occur on the same day as the take-off date, and calculating the influence factor of the external source event on the civil aviation demand of the current flight according to a preset strategy.
Optionally, before determining whether there is an abnormality in the takeoff date of the current flight according to a standard curve of civil aviation demand of the week of the airline where the current flight is located and a abnormal curve of civil aviation demand of the airline, the method further includes: analyzing a log file of a civil aviation information system and calculating the weather-level civil aviation demand data; eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the current flight path, and then averaging to obtain the week civil aviation demand standard curve; comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shapes, and reserving the central curve of each type as the civil aviation demand abnormal curve of the airline.
Optionally, predicting the prediction index of the day-level civil aviation demand of the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises: and under the condition that the prediction type is that the takeoff date of the current flight is not abnormal and the influence factor of an exogenous event to be generated on the current date of the current flight on the civil aviation demand of the airline where the current flight is located is zero, predicting the astronomical demand prediction index of the current flight on the takeoff date according to the historical inquired date civil aviation demand curve and the standard curve of the civil aviation demand.
Optionally, predicting the prediction index of the day-level civil aviation demand of the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises: and under the condition that the prediction type is that the takeoff date of the current flight is abnormal and the influence factor of an exogenous event to be generated on the current flight on the civil aviation demand of the airline where the current flight is located is zero, predicting the day-level civil aviation demand prediction index of the current flight on the takeoff date according to the historical inquired date civil aviation demand curve, the week civil aviation demand standard curve and the civil aviation demand abnormal curve.
Optionally, predicting the prediction index of the day-level civil aviation demand of the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises: and when the prediction type is that the takeoff date of the current flight is not abnormal and the influence factor of an exogenous event to be generated on the current flight on the takeoff date of the current flight on the civil aviation demand of the airline where the current flight is located is not zero, overlapping the influence factor to the standard curve of the civil aviation demand on the week, and predicting the prediction index of the civil aviation demand of the current flight on the takeoff date by combining the curve of the civil aviation demand on the historical inquired date.
Optionally, predicting the prediction index of the day-level civil aviation demand of the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises: under the condition that the prediction type is that the takeoff date of the current flight is abnormal and the influence factor of an exogenous event to be generated on the current flight on the civil aviation demand of the airline where the current flight is located on the current date is not zero, predicting a first-day civil aviation demand prediction index of the current flight on the takeoff date according to the historical inquired date civil aviation demand curve, the weekly civil aviation demand standard curve and the civil aviation demand abnormal curve; superposing the influence factors to the standard curve of the civil aviation demand on the week, and predicting the prediction index of the civil aviation demand of the current flight on the second day of the takeoff date by combining the curve of the civil aviation demand of the historical inquired date; and determining the space-level civil aviation demand prediction index of the current flight on the takeoff date according to the first space-level civil aviation demand prediction index and the second space-level civil aviation demand prediction index.
According to another aspect of the present invention, there is also provided an event-based civil aviation demand prediction apparatus, including: the first determining module is used for determining the prediction type of the current flight on the takeoff date according to the departure date of the current flight and the influence factors of the exogenous events to be generated on the current date of the departure date on the civil aviation demand of the airline where the current flight is located; and the prediction module is used for predicting the day-level civil aviation demand prediction index of the current flight on the take-off date according to the prediction type of the current flight on the take-off date.
Optionally, the apparatus further comprises: the second determining module is used for determining whether the takeoff date of the current flight is abnormal or not according to the standard curve of civil aviation demand of the route where the current flight is located and the abnormal curve of civil aviation demand of the route; and the first calculation module is used for acquiring an external event which will occur on the same day as the take-off date and calculating the influence factor of the external event on the civil aviation demand of the current flight according to a preset strategy.
Optionally, the apparatus further comprises: the second calculation module is used for analyzing the log file of the civil aviation information system and calculating the weather-level civil aviation demand data; the third calculation module is used for eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the route where the current flight is located and then averaging to obtain the week-level civil aviation demand standard curve; and the fourth calculation module is used for comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shape, and reserving the central curve of each type as the civil aviation demand abnormal curve of the airline.
According to the method, the prediction type of the current flight on the take-off date is determined by adopting the influence factors of whether the take-off date of the current flight is abnormal or not and the external source events which will occur on the current day on the civil aviation demand of the current flight on the current flight line on the take-off date; according to the prediction type of the current flight on the take-off date, the method for predicting the weather-level civil aviation demand prediction index of the current flight on the take-off date solves the problem of inaccurate civil aviation demand prediction, adds the influence of an external event on the civil aviation demand, and improves the accuracy of the civil aviation demand prediction.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of a method for forecasting civil aviation needs according to an embodiment of the invention;
FIG. 2 is a block diagram of an event-based civil aviation demand prediction apparatus according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a civil aviation demand index prediction model according to a preferred embodiment of the invention;
FIG. 4 is a functional block diagram of a civil aviation demand index prediction model in accordance with a preferred embodiment of the present invention;
FIG. 5 is an architecture diagram of a civil aviation demand index prediction model in accordance with a preferred embodiment of the present invention;
FIG. 6 is a flow diagram of anomaly discovery in accordance with a preferred embodiment of the present invention;
FIG. 7 is a flowchart of a process for logically classifying according to prediction type in accordance with a preferred embodiment of the present invention.
Detailed Description
The invention will be described in detail hereinafter with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
Example 1
In this embodiment, a method for predicting civil aviation demand is provided, and fig. 1 is a flowchart of the method for predicting civil aviation demand according to the embodiment of the present invention, as shown in fig. 1, the process includes the following steps:
step S101, determining the prediction type of the current flight on the take-off date according to the take-off date of the current flight and the influence factors of the exogenous events to be generated on the current date on the civil aviation demand of the current flight;
and S102, predicting the day-level civil aviation demand prediction index of the current flight on the take-off date according to the prediction type of the current flight on the take-off date.
Through the steps, when the prediction of the weather-level civil aviation demand prediction index is carried out, the influence factor of the exogenous event on the civil aviation demand is considered, so that the influence of the exogenous event can be sensed by the civil aviation demand prediction index, and the accuracy of the civil aviation demand prediction is improved. The weather-level civil aviation demand prediction indexes of multiple continuous days in the future are counted, a civil aviation demand prediction index curve of multiple days in the future can be obtained, and the change trend of the civil aviation demand prediction indexes is displayed.
Optionally, before determining the predicted type of the current flight on the takeoff date according to whether there is an abnormality in the takeoff date of the current flight and whether there is an event affecting the civil aviation demand on the same date as the takeoff date, the method further includes: determining whether the departure date of the current flight is abnormal or not according to the standard curve of the civil aviation demand of the current flight on the airline and the abnormal curve of the civil aviation demand of the airline; and acquiring an external source event which will occur on the same day as the take-off date, and calculating the influence factor of the external source event on the civil aviation demand of the current flight on the airline according to a preset strategy.
Optionally, the standard curve of the demand of the civil aviation on the week is generated by the following method: analyzing a log file of a civil aviation information system and calculating the weather-level civil aviation demand data; and eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the current flight line, and then averaging to obtain a week civil aviation demand standard curve.
Optionally, the civil aviation demand anomaly curve is obtained by: comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shape, and reserving the central curve of each type as the civil aviation demand abnormal curve of the airline.
Optionally, the prediction type according to the current flight on the departure date comprises four types, namely: no abnormal event, and abnormal event. And when the prediction index of the daily civil aviation demand is predicted, different prediction logics or prediction models are used for prediction according to different prediction types.
For example, when the prediction type is that the takeoff date of the current flight is not abnormal and the influence factor of an exogenous event which will occur on the current flight on the civil aviation demand of the airline where the current flight is located is zero (namely no event is considered), the day-level civil aviation demand prediction index of the current flight on the takeoff date is predicted according to the historical inquired date civil aviation demand curve and the week civil aviation demand standard curve.
For example, when the predicted type is that the takeoff date of the current flight is abnormal and the influence factor of an exogenous event which will occur on the current flight on the civil aviation demand of the airline where the current flight is located is zero (namely no event is considered), the day-level civil aviation demand prediction index of the current flight on the takeoff date is predicted according to the historical inquired date civil aviation demand curve, the week civil aviation demand standard curve and the civil aviation demand abnormal curve.
For example, when the prediction type is that the takeoff date of the current flight is not abnormal and the influence factor of an exogenous event which will occur on the current flight departure date on the civil aviation demand of the airline where the current flight is located is not zero, the influence factor is superposed on the standard curve of the civil aviation demand in week, and the prediction index of the civil aviation demand of the current flight on the takeoff date in day is predicted by combining the curve of the civil aviation demand in the historical inquired date.
For example, under the condition that the prediction type is that the takeoff date of the current flight is abnormal and the influence factor of an exogenous event which will occur on the current flight on the civil aviation demand of the airline where the current flight is located is not zero, predicting the first-level civil aviation demand prediction index of the current flight on the takeoff date according to a historical inquired date civil aviation demand curve, a weekly civil aviation demand standard curve and a civil aviation demand abnormal curve; the influence factors are superposed on the standard curve of the civil aviation demand in the week, and the prediction index of the civil aviation demand of the second day of the takeoff date of the current flight is predicted by combining the historical inquired date civil aviation demand curve; and determining the day-level civil aviation demand prediction index of the current flight on the takeoff date according to the first day-level civil aviation demand prediction index and the second day-level civil aviation demand prediction index.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
Example 2
In this embodiment, an event-based civil aviation demand prediction apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and the description of which has been already made is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 2 is a block diagram of an event-based civil aviation demand prediction apparatus according to an embodiment of the present invention, as shown in fig. 2, the apparatus includes: a first determination module 21 and a prediction module 22, wherein,
the first determining module 21 is configured to determine a prediction type of the current flight on the take-off date according to whether the take-off date of the current flight is abnormal or not and an influence factor of an external event that will occur on the current date on a civil aviation demand of an airline where the current flight is located; and the predicting module 22 is coupled to the first determining module 21 and configured to predict the prediction index of the day-level civil aviation demand of the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date.
Optionally, the apparatus further comprises: the second determining module is coupled to the first determining module 21 and is used for determining whether the departure date of the current flight is abnormal or not according to the standard curve of the civil aviation demand of the route where the current flight is located and the abnormal curve of the civil aviation demand of the route; and the first calculating module is coupled to the first determining module 21 and is used for acquiring an external event which will occur on the same day as the takeoff date and calculating an influence factor of the external event on the civil aviation demand of the current flight according to a preset strategy.
Optionally, the apparatus further comprises: the second calculation module is used for analyzing the log file of the civil aviation information system and calculating the weather-level civil aviation demand data; the third calculation module is coupled to the second calculation module and the second determination module and used for eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the route where the current flight is located and then averaging the day-level civil aviation demand data to obtain a standard curve of the week civil aviation demand; and the fourth calculation module is coupled to the third calculation module and the second determination module and is used for comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shape, and reserving the central curve of each type as the civil aviation demand abnormal curve of the airline.
Optionally, the prediction module 22 is configured to, when the prediction type is that the takeoff date of the current flight is not abnormal and an influence factor of an external event that will occur on the current date of the current flight on the civil aviation demand of the airline where the current flight is located is zero, predict an day-level civil aviation demand prediction index of the current flight on the takeoff date according to a historical queried date civil aviation demand curve and a week civil aviation demand standard curve.
Optionally, the prediction module 22 is configured to, when the prediction type is that the takeoff date of the current flight is abnormal and an influence factor of an external event that will occur on the current date of the current flight on the civil aviation demand of the airline where the current flight is located is zero, predict an antenna-level civil aviation demand prediction index of the current flight on the takeoff date according to a historical queried date civil aviation demand curve, a weekly civil aviation demand standard curve and a civil aviation demand abnormal curve.
Optionally, the prediction module 22 is configured to, when the prediction type is that the takeoff date of the current flight is not abnormal, and an influence factor of an external event that will occur on the current flight on the civil aviation demand of the airline where the current flight is located is not zero, superimpose the influence factor on the standard curve of the civil aviation demand, and predict the day-level civil aviation demand prediction index of the current flight on the takeoff date by combining with the historical queried date civil aviation demand curve.
Optionally, the prediction module 22 is configured to, when the prediction type is that the takeoff date of the current flight is abnormal and an influence factor of an external event that will occur on the current flight on the civil aviation demand of the airline where the current flight is located on the current day is not zero, predict a first-day civil aviation demand prediction index of the current flight on the takeoff date according to a historical queried-date civil aviation demand curve, a standard curve of the civil aviation demand and an abnormal curve of the civil aviation demand; the influence factors are superposed on the standard curve of the civil aviation demand in the week, and the prediction index of the civil aviation demand of the second day of the takeoff date of the current flight is predicted by combining the historical inquired date civil aviation demand curve; and determining the day-level civil aviation demand prediction index of the current flight on the takeoff date according to the first day-level civil aviation demand prediction index and the second day-level civil aviation demand prediction index.
It should be noted that, the above modules may be implemented by software or hardware, and for the latter, the following may be implemented, but not limited to: the modules are all positioned in the same processor; alternatively, the modules are respectively located in a plurality of processors.
Example 3
The embodiment of the present invention also provides software for executing the technical solutions described in the above embodiments and preferred embodiments.
Example 4
The embodiment of the invention also provides a storage medium. In the present embodiment, the storage medium described above may be configured to store program code for performing the steps of:
step S101, determining the prediction type of the current flight on the take-off date according to the take-off date of the current flight and the influence factors of the exogenous events to be generated on the current date on the civil aviation demand of the current flight;
and S102, predicting the day-level civil aviation demand prediction index of the current flight on the take-off date according to the prediction type of the current flight on the take-off date.
Optionally, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.
Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation manners, and this embodiment is not described herein again.
In order that the description of the embodiments of the invention will be more apparent, reference is now made to the preferred embodiments for illustration.
Example 5
In order to provide more intelligent demand prediction support for the revenue management of an airline company, the embodiment provides a civil aviation demand index prediction model, full query data of each channel is collected based on a large data platform, the data source range is wide, the market perceptibility is strong, anonymous in-line data of civil aviation passengers and user query log data are used as data sources, and the model mainly relates to core modules such as abnormal discovery, event and demand index correlation, demand index prediction and the like. And (4) calculating abnormal values by analyzing the standard curve on the basis of the calculated takeoff date and record date demand index curve, and classifying abnormal takeoff dates so as to realize abnormal discovery. And (3) calculating the correlation between the event and the demand index, and inputting the exogenous event into a model and calculating to obtain the influence degree of the event on the future demand index by constructing an event and demand correlation model. The demand index prediction firstly judges whether the takeoff date is abnormal or not and whether the takeoff date is an event or not, and then the data such as the standard curve library, the abnormal curve library, the event influence and the like are used for prediction according to classification, so that the day-level prediction demand index is obtained.
In order to achieve the above purpose, the civil aviation demand index prediction model provided in this embodiment obtains demand index day-level data from a log of an avigation AVE system, and stores the data into a Hive table, where the content of the data includes flight departure date, departure place, destination, user query time, and the like. And inputting the exogenous event library into a social event and demand index correlation model, and calculating to obtain event influence, wherein the model is independent off-line operation. The main system is characterized in that a week standard curve is initialized by adopting demand index data, an abnormal curve is stored in the Hive table, and a small window abnormality of a takeoff date curve is found by an abnormality finding algorithm and stored in the Hive table. And then, demand index prediction is carried out by adopting demand index data, standard curve data and abnormal curve data, and the standard curve library and the abnormal curve library can be updated in an iterative manner at regular intervals.
Fig. 3 is a schematic structural diagram of a civil aviation demand index prediction model according to a preferred embodiment of the present invention, and as shown in fig. 3, the civil aviation demand index prediction model includes an external event library developed by Java on a single machine, a demand index result model developed by Spark on a Hadoop platform, a Hive database, an event and demand correlation model, a demand index prediction model, and a prediction result output/display module.
Fig. 4 is a functional block diagram of a civil aviation demand index prediction model according to a preferred embodiment of the present invention, and fig. 5 is an architecture diagram of the civil aviation demand index prediction model according to the preferred embodiment of the present invention. Referring to fig. 4 and 5, the processing flow of the civil aviation demand index prediction model provided in this embodiment includes the following steps and links:
preprocessing data: and calculating the day-level demand index, and outputting and storing the day-level demand index into a Hive table of Hadoop according to a certain format.
Establishing and updating a standard curve library: and processing the demand index data of the same week attribute of different routes by adopting the demand index data, eliminating abnormal points, averaging to obtain a week standard curve of each route, and storing the week standard curve in a standard curve library. And determining whether the corresponding recording date curve is normal or not according to the takeoff date label, locally updating the week standard curve, if the week standard curve does not exist, adding the week standard curve, and storing the updating or adding result into the Hive table.
Establishing and updating an abnormal curve library: and for the inquired date curve of a plurality of days (for example, 80 days) corresponding to the newly added takeoff date, comparing the inquired date curve with the week standard curve corresponding to the takeoff date, performing K-means clustering on the curve set with abnormal shapes, and storing the central curve of each type into an abnormal curve library. And comparing the inquired date curve and the week standard curve of a certain product set, if the abnormal condition is presented, adding the inquired date curve of the product set into an abnormal curve library, and storing the result into the Hive table.
Event and demand correlation calculation: and constructing an event and demand correlation model, inputting an exogenous event into the model, and calculating to obtain the influence of the event on a future demand index.
Fifthly, discovering the abnormality: it can be found whether there are anomalous times and anomalous events near the departure date that have an effect on airline requirements. And (4) finding the abnormal state of the small window of the takeoff date curve and storing the abnormal state into the Hive table by adopting the demand index data, the takeoff date standard curve and the recorded date standard curve. Referring to fig. 6, in the abnormality finding, the abnormality is classified by the classification decision tree by calculating the departure date curve small window abnormal value, the recording date curve overall morphology abnormal value, and the recording date curve abnormal value.
Referring to fig. 7, the demand index prediction module first determines whether there is an abnormality in the takeoff date and whether there is an event, and then predicts according to the different types by using data such as the standard curve library, the abnormal curve library, and the event influence, so as to obtain the daily prediction demand index.
The preferred embodiments are described and illustrated below by specific examples.
The system for predicting the civil aviation demand index provided by the embodiment comprises a data preprocessing module, a standard curve library establishing module, an abnormal curve library establishing module, an event and demand correlation module, an abnormal discovery algorithm module, a demand index predicting module, a standard curve library updating module and an abnormal curve library updating module. The function of each module is described in table 1.
TABLE 1 System Module description
According to the model for predicting the civil aviation demand index, firstly, a takeoff date curve of a current date in the future of 59 days and a historical 80-day recording date curve corresponding to each takeoff date are determined and analyzed; then determining an abnormal type label of a certain takeoff date according to an abnormal finding algorithm, moving the current date, generating a plurality of abnormal type labels of the recorded date curve, judging whether the recorded date curve is abnormal or not, and establishing a standard curve library and an abnormal curve library; and finally, adjusting the standard curve, the abnormal influence, the event influence and the like according to the demand index prediction models of different categories, and synthesizing the queried date curve to generate a final demand index prediction curve.
As shown in fig. 3 and 5, the prediction model for civil aviation demand index proposed in this example is used to implement the following steps:
step 1, data preprocessing: and calculating the day-level demand index, and outputting and storing the day-level demand index into a Hive table of Hadoop according to a certain format. The table structure is shown in table 2.
Table 2 storage structure
And 2, establishing and updating a standard curve library. And establishing a standard curve library, processing the demand index data of the same week attribute of different routes by adopting the demand index data, eliminating abnormal points, averaging to obtain a week standard curve of each route, and storing the week standard curve in the standard curve library. For example, in one embodiment, the algorithm generates a queried date curve from demand index data of 2016 month 1 to 2016 month 5, traversing the takeoff date. In order to ensure that the number of days of the inquired date curve is 80, the takeoff date range is set between 2016 (3 month and 21 days) and 2016 (6 month and 1 day) and outliers of the inquired date curves with the same week attribute of the same airline are removed to obtain an average value, so that a 7-week standard curve of each airline is obtained, and the specific implementation process is as follows:
a. acquiring queried date curves of all airline takeoff dates within a range from 2016, 3, 21 and 6, 1;
b. converting the takeoff date attribute of the acquired queried date curve into the week attribute of the takeoff date;
c. classifying the inquired date curves according to the attributes of the routes and the weeks;
d. defining p to represent the p-th point on one curve, taking the value of 0-79, and for a certain curve, taking the sum of the demand indexes of the same position as sumIndex [ p ], and taking the sum of the points with the demand indexes of the same position different from 0 as number [ p ];
e. if number [ p ] is not equal to 0, then the position average value aveIndex [ p ] ═ sumIndex [ p ]/number [ p ], otherwise aveIndex [ p ] ═ 0;
f. reading the curves in sequence, calculating whether 80 points tempIndex [ p ] of the curves are outliers or not, defining the deviation degree of the points, comparing the calculation result (Ratio) with a week standard curve outlier threshold (ratioThreshold), and if the Ratio is greater than ratioThreshold, determining the points are outliers, and obtaining SumIndex [ p ] (SumIndex [ p ] -tempIndex [ p ], and number [ p ] (number [ p ] -1);
g. and averaging the inquired date curves with the same week attribute of the same airline after outliers are removed from the same position to obtain a week standard curve of the airline, and storing all the week standard curves of all the airlines into a Hive table.
When an 80-day query curve is newly generated, updating a corresponding week standard curve in a standard curve library according to the week corresponding to the takeoff date of the curve, wherein the process comprises the following steps: if the week standard curve exists, whether the corresponding recording date curve is normal or not is determined according to the takeoff date label of the new 80-day query curve, the week standard curve is locally updated, if the week standard curve does not exist, the curve is the corresponding week standard curve, and the updated or newly added result is stored in the Hive table. The specific process is as follows:
a. acquiring week standard curves of all airlines;
b. each takeoff date curve small window comprises 3 points, so that each takeoff date point has three abnormal labels, the first label of the point is selected as the abnormal condition tag of the takeoff date point on the current date, tag 1 is normal, and tag 0 is abnormal;
c. obtaining the abnormal condition of each takeoff date point on the takeoff date curve, and taking a historical 21-day demand index curve corresponding to each takeoff date;
d. for each product set, if the tag corresponding to the takeoff date is 1, judging whether the week standard curve exists, if so, locally updating 21 points corresponding to the week standard curve by taking the historical 21-day demand index curve corresponding to the takeoff date according to the weight, otherwise, adding the corresponding part of the week standard curve;
e. and storing the updated or newly added results of the week standard curves into the Hive table.
Step 3, establishing and updating an abnormal curve library: comparing week standard curves corresponding to dates, performing K-means clustering on the abnormal lines, and storing the central curves of each type into an abnormal library. And according to the demand index data from 2016 to 2016 and 5, traversing the takeoff date to generate a queried date curve. In order to ensure that the number of days of the inquired date curve is 80, setting the takeoff date range between 2016 (3 months) and 21 days and 2016 (6 months) and 1 day, comparing the inquired date curve with the week standard curve corresponding to the takeoff date of the inquired date curve to obtain an abnormal curve, clustering the abnormal curve to obtain a class center curve, and storing the class center curve in an abnormal library, wherein the specific execution process comprises the following steps:
a. acquiring week standard curves of all airlines;
b. acquiring queried date curves of all airline takeoff dates within a range from 2016, 3, 21 and 6, 1;
c. reading the queried date curve, summing 80 points of the queried date curve to obtain sum, and filtering if the sum is less than hisSumThreshold, wherein the hisSumThreshold is the threshold value of the sum of indexes of the queried curve;
d. obtaining the week attribute corresponding to the takeoff date of each curve for the filtered inquired date curves, calculating the cosine distance CosValue between the inquired date and the week standard curve, and recording as an abnormal curve if the CosValue is greater than cosThreshold, wherein the cosThreshold is a threshold value for judging the cosine distance abnormality;
e. and setting the total number of the types of the abnormal curves as typeCount, clustering the abnormal curves by using a Kmeans algorithm to obtain typeCount types, taking a curve at the center position of the typeCount types, adding the curve into an abnormal curve library and storing the curve in a Hive table.
Comparing the inquired date curve and the week standard curve of a certain product set, if the abnormal condition is presented, adding the abnormal condition library, and storing the result into the Hive table, wherein the specific execution process is as follows:
a. acquiring all old abnormal curves and newly generated queried date curves;
b. calculating the week attribute corresponding to the takeoff date of the inquired date curve to obtain the cosine distance CosValue between the inquired date curve and the week standard curveHC,SCIf CosValueHC,SC<Filtering cosThreshold;
c. calculating the cosine distance CosValue between the rest inquired date curve and the curve in the abnormal curve libraryHC,ECIf CosValueHC,EC>Adding cosThreshold into an abnormal curve library;
d. the results are saved to the Hive table.
And 4, calculating the correlation between the event and the demand: and constructing an event and demand correlation model, inputting an exogenous event into the model, and calculating to obtain the influence of the event on a future demand index.
Step 5, abnormal finding: and (4) finding the abnormal state of the small window of the takeoff date curve and storing the abnormal state into the Hive table by adopting the demand index data, the takeoff date standard curve and the recorded date standard curve. The abnormal values are classified through a classification decision tree by calculating the abnormal values of the small windows of the takeoff date curve, the abnormal values of the integral form of the recording date curve and the abnormal values of the recording date curve, and the specific abnormal values are described, executed and decided as follows.
a. Calculating a small window abnormal value of a takeoff date curve:
when the abnormal value in the small window of the takeoff date curve is calculated, the comparison condition of the current time curve and the standard curve of the flight line is considered, the simultaneous query condition of the small window and other flight lines is compared, and the overall fluctuation condition of the current time curve is comprehensively considered. The physical distance between the current curve and the standard curve is compared through the Euclidean distance, and the form difference between the current curve and other routes is compared through the cosine distance. Calculating an abnormal value and dividing the abnormal value by the 'standard deviation of the whole curve minus the standard curve', wherein the smaller the fluctuation of the whole curve is, the more prominent the abnormal condition in the small window is; secondly, normalizing curves of different flight paths with different orders of magnitude to be within an abnormal value range. The specific calculation formula of the abnormal value of the small window of the takeoff date is as follows:
fdxOD,i∈ { fd | the current time queries the ith small window curve of takeoff date for all ODs except that OD }
In the above-mentioned formula, the compound of formula,
Sgn(fdOD,i-standardfdOD,i) And the symbol represents the sum of the ith small window of a certain OD take-off date curve minus the sum of the ith small window of the OD take-off date standard curve.
Eud(fdOD,i,standardfdOD,i) And the Euclidean distance between the ith small window of a certain OD take-off date curve and the ith small window of the OD take-off date standard curve is represented.
Ave[Cosine(fdOD,i,fdxOD,i)]And the mean value of cosine distances between the ith small window of a certain OD take-off date curve and the ith small window of all other OD take-off date curves is represented.
StdDev(fdOD-standardfdOD) And the standard deviation of a certain OD take-off date curve minus the OD take-off date standard curve is shown.
b. Calculating the abnormal value of the overall shape of the recording date curve:
the integral form of the recording date curve reflects the difference between the recording date curve and the standard curve form, and the specific calculation formula is as follows:
in the above formula, Cosine (rd)OD,i,standardrdOD,i) And the cosine distance between the recording date curve corresponding to the ith small window of the OD take-off date curve and the recording date standard curve of the small window is represented.
c. Calculating a small window abnormal value of the recording date curve:
the abnormal value of the small window of the recording date curve reflects the abnormal condition in a certain window of the recording date curve, the window with the largest difference between the current recording date curve and the standard curve is found out, the morphological difference between the window and other routes is comprehensively compared, and the abnormal value is obtained by comprehensively considering the overall fluctuation condition of the curve.
fdxOD,i,k∈ { rd | the current time queries all OD takeoff date curves except the OD corresponding to the ith small window
Kth small window of recording date curve }
In the above-mentioned formula, the compound of formula,
max[Eud(rdOD,i,standardrdOD,i)]and the maximum value of the Euclidean distance between a recording date curve corresponding to the ith small window of a certain OD take-off date curve and a recording date standard curve of the small window is shown.
Ave[Cosine(rdOD,i,k,rdxOD,i,k)]A kth small window representing a record date curve corresponding to an ith small window of a certain OD take-off date curve(the Euclidean distance of the small window is the largest compared with the standard curve) and the cosine distance of the kth small window of the recording date curve corresponding to the ith small window of all other OD takeoff date curves.
StdDev(rdOD,i-standardrdOD,i) And (3) the standard deviation of the recorded date curve of the small window subtracted from the recorded date standard curve of the small window corresponding to the ith small window representing the OD takeoff date curve.
d. The anomaly discovery and the anomaly classification are further completed by calculating the anomaly value and combining the anomaly classification method shown in FIG. 6.
And 6, forecasting a demand index: time series prediction generally needs to consider four aspects: long term trends, seasonal variations, periodic variations, and irregular variations. The first three terms can be obtained by conventional prediction methods, and the irregular variation is obtained by analyzing anomalies and events: converting window abnormity into point abnormity according to a previous abnormity discovery algorithm, and judging whether a certain takeoff date point is abnormal or not; searching in an external source event library, judging whether an event exists in a certain city on a certain date, and obtaining the influence of the event on the demand by constructing an event and demand correlation model. And (4) carrying out classified modeling according to the existence of the abnormality and the existence of the event, and predicting according to the classification method of the prediction model shown in the figure 7 by utilizing data such as a standard curve library, an abnormal curve library, event influence and the like according to the classification to obtain the day-level prediction demand index. The principle that this prediction process needs to follow is as follows:
according to whether the predicted takeoff date is abnormal or not, predicting by referring to historical curve data of a standard curve library or an abnormal curve library;
and adding the influence data of the event into a prediction model for prediction if the event exists according to whether the predicted takeoff date has the event.
In summary, through the embodiments of the present invention, effective prediction of flight requirements of passengers in the future can be achieved, historical flight query requirements of passengers are used as input, the change trends of flight requirements of passengers are classified, an abnormal curve is found, and the change trend of the requirements of each takeoff date in a period of time in the future is obtained by combining the influence of events on the requirements. Through model calculation, a prediction result can be rapidly and accurately output, and the accuracy of prediction is improved by adding event factors.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. An event-based civil aviation demand prediction method is characterized by comprising the following steps:
analyzing a log file of a civil aviation information system and calculating the weather-level civil aviation demand data;
eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the current flight line, and then averaging to obtain a week-level civil aviation demand standard curve;
the standard curve of the demands of the civil aviation on the week is generated by the following method: analyzing a log file of a civil aviation information system and calculating the weather-level civil aviation demand data; eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the current flight line, and then averaging to obtain the week civil aviation demand standard curve;
comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shape, and reserving the central curve of each type as the civil aviation demand abnormal curve of the airline;
the civil aviation demand abnormity curve is obtained by the following method: comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shape, and reserving the central curve of each type as the airline civil aviation demand abnormal curve;
determining whether the departure date of the current flight is abnormal or not according to the standard curve of the civil aviation demand of the current flight on the airline and the abnormal curve of the civil aviation demand of the airline;
acquiring an external source event which will occur on the same day as the take-off date, and calculating an influence factor of the external source event on the civil aviation demand of the current flight according to a preset strategy;
determining the prediction type of the current flight on the take-off date according to the take-off date of the current flight is abnormal and the influence factor of an exogenous event to be generated on the current date on the civil aviation demand of the current flight on the route of the current flight;
and predicting the day-level civil aviation demand prediction index of the current flight on the take-off date according to the prediction type of the current flight on the take-off date.
2. The method of claim 1, wherein predicting the prediction index of the day-level civil aviation demand for the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises:
and under the condition that the prediction type is that the takeoff date of the current flight is not abnormal and the influence factor of an exogenous event to be generated on the current date of the current flight on the civil aviation demand of the airline where the current flight is located is zero, predicting the astronomical demand prediction index of the current flight on the takeoff date according to the historical inquired date civil aviation demand curve and the standard curve of the civil aviation demand.
3. The method of claim 1, wherein predicting the prediction index of the day-level civil aviation demand for the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises:
and under the condition that the prediction type is that the takeoff date of the current flight is abnormal and the influence factor of an exogenous event to be generated on the current flight on the civil aviation demand of the airline where the current flight is located is zero, predicting the day-level civil aviation demand prediction index of the current flight on the takeoff date according to the historical inquired date civil aviation demand curve, the week civil aviation demand standard curve and the civil aviation demand abnormal curve.
4. The method of claim 1, wherein predicting the prediction index of the day-level civil aviation demand for the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises:
and when the prediction type is that the takeoff date of the current flight is not abnormal and the influence factor of an exogenous event to be generated on the current flight on the takeoff date of the current flight on the civil aviation demand of the airline where the current flight is located is not zero, overlapping the influence factor to the standard curve of the civil aviation demand on the week, and predicting the prediction index of the civil aviation demand of the current flight on the takeoff date by combining the curve of the civil aviation demand on the historical inquired date.
5. The method of claim 1, wherein predicting the prediction index of the day-level civil aviation demand for the current flight on the takeoff date according to the prediction type of the current flight on the takeoff date comprises:
under the condition that the prediction type is that the takeoff date of the current flight is abnormal and the influence factor of an exogenous event to be generated on the current flight on the civil aviation demand of the airline where the current flight is located on the current date is not zero, predicting a first-day civil aviation demand prediction index of the current flight on the takeoff date according to the historical inquired date civil aviation demand curve, the weekly civil aviation demand standard curve and the civil aviation demand abnormal curve; superposing the influence factors to the standard curve of the civil aviation demand on the week, and predicting the prediction index of the civil aviation demand of the current flight on the second day of the takeoff date by combining the curve of the civil aviation demand of the historical inquired date; and determining the space-level civil aviation demand prediction index of the current flight on the takeoff date according to the first space-level civil aviation demand prediction index and the second space-level civil aviation demand prediction index.
6. Civil aviation demand forecasting device for implementing an event-based civil aviation demand forecasting method according to any one of claims 1 to 5, characterized in that it comprises:
the first determining module is used for determining the prediction type of the current flight on the takeoff date according to the departure date of the current flight and the influence factors of the exogenous events to be generated on the current date of the departure date on the civil aviation demand of the airline where the current flight is located;
the prediction module is used for predicting a day-level civil aviation demand prediction index of the current flight on the take-off date according to the prediction type of the current flight on the take-off date;
the second determining module is used for determining whether the takeoff date of the current flight is abnormal or not according to the standard curve of civil aviation demand of the route where the current flight is located and the abnormal curve of civil aviation demand of the route;
the first calculation module is used for acquiring an external event which will occur on the same day as the take-off date and calculating the influence factor of the external event on the civil aviation demand of the current flight according to a preset strategy; the second calculation module is used for analyzing the log file of the civil aviation information system and calculating the weather-level civil aviation demand data;
the third calculation module is used for eliminating abnormal points of the day-level civil aviation demand data with the same week attribute of the route where the current flight is located and then averaging to obtain the week-level civil aviation demand standard curve;
and the fourth calculation module is coupled to the third calculation module and the second determination module and is used for comparing the historical inquired date civil aviation demand curve of the airline where the current flight is located with the week civil aviation demand standard curve corresponding to the takeoff date of the current flight, performing K-means clustering on the historical inquired date civil aviation demand curve with abnormal shape, and reserving the central curve of each type as the civil aviation demand abnormal curve of the airline.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710004030.3A CN106779245B (en) | 2017-01-04 | 2017-01-04 | Event-based civil aviation demand prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710004030.3A CN106779245B (en) | 2017-01-04 | 2017-01-04 | Event-based civil aviation demand prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106779245A CN106779245A (en) | 2017-05-31 |
| CN106779245B true CN106779245B (en) | 2020-09-04 |
Family
ID=58949443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710004030.3A Active CN106779245B (en) | 2017-01-04 | 2017-01-04 | Event-based civil aviation demand prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106779245B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451750A (en) * | 2017-08-10 | 2017-12-08 | 中国民航信息网络股份有限公司 | Civil aviaton's demand anomaly method and apparatus |
| CN109598378A (en) * | 2018-11-30 | 2019-04-09 | 平安医疗健康管理股份有限公司 | Medical insurance prediction technique, device, equipment and computer readable storage medium |
| CN109711894B (en) * | 2018-12-28 | 2020-10-30 | 携程旅游网络技术（上海）有限公司 | Method and system for detecting abnormal flight path requirements |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102831487A (en) * | 2012-08-13 | 2012-12-19 | 南京莱斯信息技术股份有限公司 | Flow predicted result verification method based on historical scheduled flight running data analysis |
| CN104036446A (en) * | 2014-06-13 | 2014-09-10 | 中国民航信息网络股份有限公司 | Civil aviation information system based system and method for predicting preorder flight of duty flight |
| CN104077915A (en) * | 2014-03-27 | 2014-10-01 | 中华电信股份有限公司 | Riding trend prediction device and method |
-
2017
- 2017-01-04 CN CN201710004030.3A patent/CN106779245B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102831487A (en) * | 2012-08-13 | 2012-12-19 | 南京莱斯信息技术股份有限公司 | Flow predicted result verification method based on historical scheduled flight running data analysis |
| CN104077915A (en) * | 2014-03-27 | 2014-10-01 | 中华电信股份有限公司 | Riding trend prediction device and method |
| CN104036446A (en) * | 2014-06-13 | 2014-09-10 | 中国民航信息网络股份有限公司 | Civil aviation information system based system and method for predicting preorder flight of duty flight |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106779245A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2842085B1 (en) | Database system using batch-oriented computation | |
| CN106933956B (en) | Data mining method and device | |
| CN113888235B (en) | Training method of sales forecasting model, sales forecasting method and related device | |
| US20170109667A1 (en) | Automaton-Based Identification of Executions of a Business Process | |
| US20170109636A1 (en) | Crowd-Based Model for Identifying Executions of a Business Process | |
| Wohlfarth et al. | A data-mining approach to travel price forecasting | |
| CN115423289B (en) | Intelligent plate processing workshop data processing method and terminal | |
| CN111899059A (en) | Navigation driver revenue management dynamic pricing method based on block chain | |
| CN109492858A (en) | Employee performance prediction technique and device, equipment, medium based on machine learning | |
| CN112860769B (en) | Energy planning data management system | |
| CN106779245B (en) | Event-based civil aviation demand prediction method and device | |
| US11308102B2 (en) | Data catalog automatic generation system and data catalog automatic generation method | |
| CN117557299B (en) | Marketing planning method and system based on computer assistance | |
| CN114997916A (en) | Prediction method, system, electronic device and storage medium of potential user | |
| Saravanan et al. | Forecasting Economy using Machine Learning Algorithm | |
| CN115619571A (en) | Financing planning method, system and device | |
| US10055469B2 (en) | Method and software for retrieving information from big data systems and analyzing the retrieved data | |
| CN110751359B (en) | Automatic route network evaluation method, electronic equipment and storage medium | |
| CN111612491B (en) | State analysis model construction method, analysis method and device | |
| Wang | An intelligent passenger flow prediction method for pricing strategy and hotel operations | |
| KR20230052010A (en) | Demand forecasting method using ai-based model selector algorithm | |
| Zhang et al. | Dynamic time warp-based clustering: Application of machine learning algorithms to simulation input modelling | |
| CN116308494A (en) | Supply chain demand prediction method | |
| CN115098572A (en) | User portrait construction method and device and computer equipment | |
| Zhang et al. | Retail commodity sale forecast model based on data mining |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |