CN108846523A - A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network - Google Patents
A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network Download PDFInfo
- Publication number
- CN108846523A CN108846523A CN201810856405.3A CN201810856405A CN108846523A CN 108846523 A CN108846523 A CN 108846523A CN 201810856405 A CN201810856405 A CN 201810856405A CN 108846523 A CN108846523 A CN 108846523A
- Authority
- CN
- China
- Prior art keywords
- flight
- putting forth
- data
- coasting time
- bayesian network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network.It includes constituting air station flight departure from port operation data collection；Obtain complete air station flight departure from port operation data collection；Determine major influence factors；Obtaining Bayesian network flight for putting forth can be changed coasting time dynamic estimation model；The variable coasting time of estimation flight for putting forth and etc..Flight for putting forth coasting time dynamic prediction method provided by the invention based on Bayesian network has the following advantages that：Establishing the flight for putting forth for combining machine learning techniques to combine with civil aviaton's expertise can be changed coasting time dynamic estimation model, airdrome scene situation locating for outgoing flight can accurately, be quickly analyzed, lets pass for Civil Aviation Airport flight and management provides strong decision support.This method had both had simple, efficient, high-precision feature, had reflected flight for putting forth slides in aerodrome traffic system independence and individual difference, and analysis and assessment application value with higher to airdrome scene traffic circulation situation.
Description
Technical field
The invention belongs to computer simulation technique fields, slide more particularly to a kind of flight for putting forth based on Bayesian network
Time dynamic prediction method.
Background technique
Airport is the important pivot of air transport system, worldwide main as air transportation business amount is increased sharply
Phenomena such as all there is different degrees of capacity saturation problem in airport, also consequently leads to air station flight delay, congestion.How to lead to
It crosses the scene situation that dynamic prediction method faces flight for putting forth to be deduced, calculated and evaluated, and adopts an effective measure and mention
High airport overall operation efficiency, to reduce the difficult point that flight delay has become computer simulation technique field.
Bayesian network is a directed acyclic graph (Directed Acyclic Graphs, DAG), wherein each node all bands
There is respective stochastic variable, the directed edge between node indicates the relationship between variable.The method of building Bayesian network has following three kinds：
First is that obtaining priori knowledge by the machine learning to mass data using big data as support, i.e., determined by parameter learning
The network structure and parameter of Bayesian network；Second is that qualitatively establishing each node according to industry practitioner and the experience of expert etc.
Incidence relation and strength of association, this is highly useful one in the case where model construction initial stage and missing key link data
Kind method.Third is that both the above method is combined, i.e., it is to rely on the background knowledge of expert, tentatively establishes Bayes
Pessimistic concurrency control simultaneously qualitativelys determine model structure；Then under the support of operation big data, verifying, sophisticated model structure and quantitative
Determine the key parameter in model.But the flight for putting forth coasting time dynamic prediction side based on Bayesian network is not yet found at present
Method.
Summary of the invention
To solve the above-mentioned problems, when being slided the purpose of the present invention is to provide a kind of flight for putting forth based on Bayesian network
Between dynamic prediction method.
In order to achieve the above object, Bayesian network building provided by the invention and prediction technique include carrying out in order
The following steps：
1) according to the target airdrome scene method of operation, each field in airdrome scene operation data library is analyzed, is therefrom extracted
Flight for putting forth operation data, and air station flight departure from port operation data collection is made of these data；
2) determine whether above-mentioned air station flight departure from port operation data collection is complete, it is such as incomplete, it is obtained using maximum likelihood algorithm
Missing data is obtained, thus to obtain complete air station flight departure from port operation data collection；
3) it determines that flight for putting forth is variable according to the data that above-mentioned complete air station flight departure from port operation data is concentrated by expert to slide
Then each influence factor of row time analyzes each influence factor using factor analysis one by one, when determining aircraft gate, departure from port
Between, Departure airport deviation, to slide flight quantity and delay this 5 influence factors of situation be major influence factors, and by each shadow
The factor of sound is as a node；Then the mutual information between each influence factor is examined, to verify the independence of each node；
4) Bayesian network estimation model is built using Netica, is then introduced into complete air station flight departure from port operation data and concentrates
The data of above-mentioned 5 influence factors carry out parameter learning, thus the Bayesian network flight under the conditions of obtaining different affecting factors from
Port can be changed coasting time and estimate model, and the newly-increased data of each influence factor obtained in the recent period are added to complete data later and are concentrated
And coasting time estimation model can be changed to flight for putting forth using these data and carry out incremental learning, obtain Bayesian network flight for putting forth
Variable coasting time dynamic estimation model；
5) actual operating data of 5 influence factors of flight to be predicted is inputted above-mentioned Bayesian network flight for putting forth can be changed
In coasting time dynamic estimation model, estimating the flight for putting forth can be changed the Posterior probability distribution of coasting time, gives up and wherein sends out
The raw lesser value item of probability, then can be changed coasting time value T using flight for putting forthiAnd its Probability p (i) calculates desired value
E (n), this desired value are that the flight for putting forth can be changed coasting time.
In step 1), the air station flight departure from port operation data is concentrated the execution date for having recorded flight in detail, is gone out
Port exception, abnormal cause, aircraft gate and plan expect, practical departure time, remove the wheel shelves moment, Departure airport, Departure airport
Deviation, the data for sliding flight quantity and being delayed including situation.
In step 2), the realization process of the maximum likelihood algorithm is as follows：
Step 1, when carrying out first time iteration, first select one group of primary data θ identical with missing data0As repairing
Data obtain broken power sample, then calculate the first generation maximum likelihood estimator θ based on repairing data1, formula is as follows：
WhereinIt is the sum of the weight for meeting condition sample in repairing data；rlIndicate the number for meeting condition sample,
I.e. based on the maximum likelihood estimator of repairing data；
Step 2 continues iteration, and according to its maximum likelihood estimator θ in the t times iterationtIt is desired right to obtain
Number likelihood function Q (θ | θt)；
Step 3, ask make log-likelihood function Q obtained in step 2 (θ | θt) maximum likelihood estimator when reaching maximumθ,
And maximum likelihood estimator θ when calculating the t+1 times iteration according to the following formulat+1：
Step 4, repeat step 2, step 3 until maximum likelihood function restrain when, algorithm terminates, i.e., when utilize maximum likelihood
Log-likelihood function sequence { l (θ obtained by algorithm0| D), l (θ1| D), l (θ2| D) ..., l (θt| D), l (θt+1| D) } in the t times
The maximum likelihood estimator θ obtained with the t+1 times iterationtWith θt+1Meet algorithmic statement condition l (θt|D)≤l(θt+1| D) when, it is defeated
Maximum likelihood estimator θ when the t+1 times iteration outt+1As missing data.
In step 3), the formula of the inspection mutual information is：
X in formulai, XjIndicate the stochastic variable of two node on behalf, P (Xi, Xj) indicate stochastic variable Xi, XjJoint Distribution it is general
Rate, P (Xi)P(Xj) respectively indicate stochastic variable Xi, XjProbability；
Successively calculate the mutual mutual information M (X of 5 influence factorsi, Xj), variable threshold ε is set as 0.2, as M (Xi,
Xj) < ε when, illustrate that two influence factors are independent mutually.
In step 5), described can be changed coasting time value T using flight for putting forthiAnd its Probability p (i) calculates expectation
The formula of value E (n) is：
E (n)=∑ (p (i) × Ti)/∑p(i) (4)。
Flight for putting forth coasting time dynamic prediction method provided by the invention based on Bayesian network has the following advantages that：It builds
Having found the flight for putting forth combined in conjunction with machine learning techniques with civil aviaton's expertise can be changed coasting time dynamic estimation model, energy
Accurately, airdrome scene situation locating for outgoing flight is quickly analyzed, lets pass for Civil Aviation Airport flight and provides strong determine with management
Plan is supported.This method had both had simple, efficient, high-precision feature, and it is sliding to have reflected flight for putting forth in aerodrome traffic system
Capable independence and individual difference, and analysis to airdrome scene traffic circulation situation and assessment is with higher applies valence
Value.
Detailed description of the invention
Fig. 1 is the flight for putting forth coasting time dynamic prediction method flow chart provided by the invention based on Bayesian network.
Fig. 2 is the average departure from port coasting time comparison diagram of different aircraft gates.
Fig. 3 is the average departure from port coasting time comparison diagram of different Departure airport.
Fig. 4 is the average departure from port coasting time comparison diagram of different Departure airport deviations.
Fig. 5 is the data learning outcome that Bayesian network estimates model.
Fig. 6 is Posterior probability distribution under the conditions of known aircraft gate and plan Departure airport.
Fig. 7 is the Posterior probability distribution under known priori conditions.
Specific embodiment
When being slided in the following with reference to the drawings and specific embodiments to the flight for putting forth provided by the invention based on Bayesian network
Between dynamic prediction method be described in detail.
As shown in Figure 1, the flight for putting forth coasting time dynamic prediction method packet provided by the invention based on Bayesian network
Include the following steps carried out in order：
1) according to the target airdrome scene method of operation, each field in airdrome scene operation data library is analyzed, is therefrom extracted
Flight for putting forth operation data, and air station flight departure from port operation data collection is made of these data；Air station flight departure from port operation data
The operative flight departure from port operation data example of concentration is as shown in table 1：
Table 1, flight for putting forth operation data example
Air station flight departure from port operation data concentration has recorded the execution date of flight, exception of leaving the port, abnormal cause, stops in detail
Seat in the plane and plan expect, practical departure time, remove the wheel shelves moment, the Departure airport, Departure airport deviation, slide flight quantity
With the data including delay situation.Wherein the Proposed Departure moment corresponds to flight planning, it is contemplated that departure time is corresponding according to practical feelings
Condition revised departure time.
2) determine whether above-mentioned air station flight departure from port operation data collection is complete, it is such as incomplete, it is calculated using maximum likelihood (EM)
Method obtains missing data, thus to obtain complete air station flight departure from port operation data collection；
The realization process of maximum likelihood algorithm is as follows：
Step 1, when carrying out first time iteration, first select one group of primary data θ identical with missing data0As repairing
Data obtain broken power sample, then calculate the first generation maximum likelihood estimator θ based on repairing data1, formula is as follows：
WhereinIt is the sum of the weight for meeting condition sample in repairing data；rlIndicate the number for meeting condition sample,
I.e. based on the maximum likelihood estimator of repairing data；
Step 2 continues iteration, and according to its maximum likelihood estimator θ in the t times iterationtIt is desired right to obtain
Number likelihood function Q (θ | θt)；
Step 3, ask make log-likelihood function Q obtained in step 2 (θ | θt) maximum likelihood estimator when reaching maximum
θ, and maximum likelihood estimator θ when calculating the t+1 times iteration according to the following formulat+1：
Step 4, repeat step 2, step 3 until maximum likelihood function restrain when, algorithm terminates, i.e., when utilize maximum likelihood
Log-likelihood function sequence { l (θ obtained by algorithm0| D), l (θ1| D), l (θ2| D) ..., l (θt| D), l (θt+1| D) } in the t times
The maximum likelihood estimator θ obtained with the t+1 times iterationtWith θt+1Meet algorithmic statement condition l (θt|D)≤l(θt+1| D) when, it is defeated
Maximum likelihood estimator θ when the t+1 times iteration outt+1As missing data.
3) it determines that flight for putting forth is variable according to the data that above-mentioned complete air station flight departure from port operation data is concentrated by expert to slide
Then each influence factor of row time analyzes each influence factor using factor analysis one by one, when determining aircraft gate, departure from port
Between, Departure airport deviation, to slide flight quantity and delay this 5 influence factors of situation be major influence factors, and by each shadow
The factor of sound is as a node；Wherein three influence factors to it is average departure from port coasting time influence degree as shown in figs 2-4.
Then the mutual information between each influence factor is examined using formula (3), to verify the independence of each node.
Calculate mutual information expression formula be：
X in formulai, XjIndicate the stochastic variable of two node on behalf, P (Xi, Xj) indicate stochastic variable Xi, XjJoint Distribution it is general
Rate, P (Xi)P(Xj) respectively indicate stochastic variable Xi, XjProbability.From formula (3) it is found that working as P (Xi, Xj)=P (Xi)P(Xj) when, M
(Xi, Xj)=0 indicates that two stochastic variables are completely independent mutually.M(Xi, Xj) closer 0, stochastic variable Xi, XjDegree of independence
It is higher.
But in the Bayesian network according to actual implementation, often seldom there is completely mutual independent stochastic variable.For just
In application, usually judge whether two stochastic variables are independent by the way that the method for variable threshold ε is arranged.As M (Xi, Xj) > ε when, table
Show that two stochastic variables are dependent relationship；It is on the contrary then be independence.
Successively calculate the mutual mutual information M (X of 5 influence factorsi, Xj), variable threshold ε is set as 0.2, as M (Xi,
Xj) < ε when, illustrate that two influence factors are independent mutually, can be variable to flight for putting forth from different dimensional representation scene influence factors
The influence of coasting time, calculated result are as shown in table 2.
Mutual information between table 2, each influence factor
4) Bayesian network estimation model is built using Netica, is then introduced into complete air station flight departure from port operation data and concentrates
The data of above-mentioned 5 influence factors carry out parameter learning, thus the Bayesian network flight under the conditions of obtaining different affecting factors from
Port can be changed coasting time and estimate model, and the newly-increased data of each influence factor obtained in the recent period are added to complete data later and are concentrated
And coasting time estimation model can be changed to flight for putting forth using these data and carry out incremental learning, obtain Bayesian network flight for putting forth
Variable coasting time dynamic estimation model；
Fig. 5 is can using Bayesian network flight for putting forth under the priori conditions known case such as aircraft gate and plan Departure airport
Becoming coasting time dynamic estimation model can be changed the Posterior probability distribution of coasting time estimation to flight for putting forth.From priori conditions
Know that aircraft berths seat in the plane close in right side, Departure airport TIME1, Bayesian network flight for putting forth can be changed coasting time dynamic estimation
Posterior probability distribution that model is obtained according to information above as shown in figure 5, comparison diagram 6 it is found that the Posterior probability distribution more collects
In, flight for putting forth can be changed probability of the coasting time in 15-17min and greatly improve.Fig. 7 is the posterior probability under known priori conditions
Distribution, it is seen that the Posterior probability distribution that flight for putting forth corresponding with the priori conditions can be changed coasting time is concentrated
In 12-17min.It can be seen that from the above analysis, obtain situation according to priori conditions, Bayesian network flight for putting forth can be changed coasting time
Dynamic estimation model can obtain the posterior probability of degree of correspondence and accuracy, adapt to the requirement of practical scene operation.
5) actual operating data of 5 influence factors of flight to be predicted is inputted above-mentioned Bayesian network flight for putting forth can be changed
In coasting time dynamic estimation model, estimating the flight for putting forth can be changed the Posterior probability distribution of coasting time, gives up and wherein sends out
The raw lesser value item of probability, then can be changed coasting time value T using flight for putting forthiAnd its Probability p (i) is counted according to formula (4)
It calculates desired value E (n), this desired value is that the flight for putting forth can be changed coasting time；
E (n)=∑ (p (i) × Ti)/∑p(i) (4)
The actual operating data of the present invention certain flight as input is given in table 3, the flight estimated according to input
The results are shown in Table 4 for the Posterior probability distribution of the variable coasting time of departure from port.
Table 3, certain flight actual operating data
Table 4, flight for putting forth can be changed the Posterior probability distribution result of coasting time
Give up the lesser value item " 18-20min " of wherein probability of happening and " being greater than 21min ", retains in rest interval
Flight for putting forth can be changed coasting time value, be 12.24min according to the desired value that formula (4) is calculated, with actual result 11min
As a result comparison, error 1.24min meet the tolerances of ± 2min, illustrate that the method for the present invention is able to satisfy current demand.
Claims (5)
1. a kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network, it is characterised in that：It is described based on shellfish
The flight for putting forth coasting time dynamic prediction method of Ye Si net includes the following steps carried out in order：
1) according to the target airdrome scene method of operation, each field in airdrome scene operation data library is analyzed, flight is therefrom extracted
Departure from port operation data, and air station flight departure from port operation data collection is made of these data；
2) determine whether above-mentioned air station flight departure from port operation data collection is complete, it is such as incomplete, it is lacked using maximum likelihood algorithm
Data are lost, thus to obtain complete air station flight departure from port operation data collection；
3) when determining that flight for putting forth is variable according to the data that above-mentioned complete air station flight departure from port operation data is concentrated by expert and sliding
Between each influence factor, then analyze each influence factor one by one using factor analysis, determine aircraft gate, the Departure airport, from
ETA estimated time of arrival deviation, to slide flight quantity and delay this 5 influence factors of situation be major influence factors, and by each influence factor
As a node；Then the mutual information between each influence factor is examined, to verify the independence of each node；
4) Bayesian network estimation model is built using Netica, it is above-mentioned is then introduced into complete air station flight departure from port operation data concentration
The data of 5 influence factors carry out parameter learning, so that the Bayesian network flight for putting forth under the conditions of obtaining different affecting factors can
Become coasting time and estimate model, the newly-increased data of each influence factor obtained in the recent period are added to complete data later and concentrate simultaneously benefit
Coasting time estimation model can be changed to flight for putting forth with these data and carry out incremental learning, it is variable to obtain Bayesian network flight for putting forth
Coasting time dynamic estimation model；
5) the above-mentioned Bayesian network flight for putting forth of actual operating data input of 5 influence factors of flight to be predicted can be changed and slides
In time dynamic estimation model, estimating the flight for putting forth can be changed the Posterior probability distribution of coasting time, gives up and wherein occurs generally
Then the lesser value item of rate can be changed coasting time value T using flight for putting forthiAnd its Probability p (i) calculates desired value E
(n), this desired value is that the flight for putting forth can be changed coasting time.
2. the flight for putting forth coasting time dynamic prediction method according to claim 1 based on Bayesian network, feature exist
In：In step 1), the described air station flight departure from port operation data concentrate have recorded in detail the execution date of flight, leave the port it is different
Often, abnormal cause, aircraft gate and plan, estimated, practical departure time, remove that take turns shelves moment, Departure airport, Departure airport inclined
Difference, the data for sliding flight quantity and being delayed including situation.
3. the flight for putting forth coasting time dynamic prediction method according to claim 1 based on Bayesian network, feature exist
In：In step 2), the realization process of the maximum likelihood algorithm is as follows：
Step 1, when carrying out first time iteration, first select one group of primary data θ identical with missing data0As repairing data,
Broken power sample is obtained, the first generation maximum likelihood estimator θ based on repairing data is then calculated1, formula is as follows：
WhereinIt is the sum of the weight for meeting condition sample in repairing data；rlIndicate the number for meeting condition sample,That is base
In the maximum likelihood estimator of repairing data；
Step 2 continues iteration, and according to its maximum likelihood estimator θ in the t times iterationtObtain desired log-likelihood
Function Q (θ | θt)；
Step 3, ask make log-likelihood function Q obtained in step 2 (θ | θt) maximum likelihood estimator θ when reaching maximum, and root
Maximum likelihood estimator θ when the t+1 times iteration is calculated according to following formulat+1：
Step 4, repeat step 2, step 3 until maximum likelihood function restrain when, algorithm terminates, i.e., when utilize maximum likelihood algorithm
Gained log-likelihood function sequence { l (θ0| D), l (θ1| D), l (θ2|D),...,l(θt|D),l(θt+1| D) } in the t times and t
The maximum likelihood estimator θ that+1 iteration obtainstWith θt+1Meet algorithmic statement condition l (θt|D)≤l(θt+1| D) when, export t
Maximum likelihood estimator θ when+1 iterationt+1As missing data.
4. the flight for putting forth coasting time dynamic prediction method according to claim 1 based on Bayesian network, feature exist
In：In step 3), the formula of the inspection mutual information is：
X in formulai, XjIndicate the stochastic variable of two node on behalf, P (Xi, Xj) indicate stochastic variable Xi, XjJoint Distribution probability, P
(Xi)P(Xj) respectively indicate stochastic variable Xi, XjProbability；
Successively calculate the mutual mutual information M (X of 5 influence factorsi, Xj), variable threshold ε is set as 0.2, as M (Xi, Xj) <
When ε, illustrate that two influence factors are independent mutually.
5. the flight for putting forth coasting time dynamic prediction method according to claim 1 based on Bayesian network, feature exist
In：In step 5), described can be changed coasting time value T using flight for putting forthiAnd its Probability p (i) calculates desired value E
(n) formula is：
E (n)=∑ (p (i) × Ti)/∑p(i) (4)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810856405.3A CN108846523A (en) | 2018-07-31 | 2018-07-31 | A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810856405.3A CN108846523A (en) | 2018-07-31 | 2018-07-31 | A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108846523A true CN108846523A (en) | 2018-11-20 |
Family
ID=64195256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810856405.3A Pending CN108846523A (en) | 2018-07-31 | 2018-07-31 | A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108846523A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807558A (en) * | 2019-11-06 | 2020-02-18 | 深圳微品致远信息科技有限公司 | Method and device for predicting departure taxi time based on deep neural network |
| CN110826788A (en) * | 2019-10-30 | 2020-02-21 | 南京智慧航空研究院有限公司 | Airport scene variable slide-out time prediction method based on big data deep learning |
| CN110852497A (en) * | 2019-10-30 | 2020-02-28 | 南京智慧航空研究院有限公司 | Scene variable slide-out time prediction system based on big data deep learning |
| CN111552178A (en) * | 2020-04-23 | 2020-08-18 | 桂林电子科技大学 | Method for controlling waiting release of aircraft stand with controllable repeat request time interval |
| CN112287561A (en) * | 2020-11-18 | 2021-01-29 | 杨媛媛 | Sliding path optimization index solving method based on Monte Carlo algorithm |
| CN113096449A (en) * | 2021-03-25 | 2021-07-09 | 上海行沙科技有限公司 | Flight big data-based shutdown position resource arrangement method and system |
| CN114783212A (en) * | 2022-03-29 | 2022-07-22 | 南京航空航天大学 | Method for constructing model feature set for prediction of departure taxi time of aircraft in busy airport |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996501A (en) * | 2010-09-10 | 2011-03-30 | 四川大学 | Novel solution method facing to flight delay |
| US20130346248A1 (en) * | 2012-06-22 | 2013-12-26 | BuyStand, LLC | Methods and electronic exchanges for facilitating a buyer-driven transaction |
| CN104156594A (en) * | 2014-08-11 | 2014-11-19 | 中国民航大学 | Dynamic flight station-crossing time estimation method based on Bayes network |
-
2018
- 2018-07-31 CN CN201810856405.3A patent/CN108846523A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996501A (en) * | 2010-09-10 | 2011-03-30 | 四川大学 | Novel solution method facing to flight delay |
| US20130346248A1 (en) * | 2012-06-22 | 2013-12-26 | BuyStand, LLC | Methods and electronic exchanges for facilitating a buyer-driven transaction |
| CN104156594A (en) * | 2014-08-11 | 2014-11-19 | 中国民航大学 | Dynamic flight station-crossing time estimation method based on Bayes network |
Non-Patent Citations (1)
| Title |
|---|
| 邢志伟、蒋骏贤、罗晓、罗谦: ""基于贝叶斯网的离港航班滑行时间动态估计"", 《计算机工程与应用》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826788A (en) * | 2019-10-30 | 2020-02-21 | 南京智慧航空研究院有限公司 | Airport scene variable slide-out time prediction method based on big data deep learning |
| CN110852497A (en) * | 2019-10-30 | 2020-02-28 | 南京智慧航空研究院有限公司 | Scene variable slide-out time prediction system based on big data deep learning |
| WO2021082393A1 (en) * | 2019-10-30 | 2021-05-06 | 南京智慧航空研究院有限公司 | Airport surface variable slide-out time prediction method based on big data deep learning |
| CN110807558A (en) * | 2019-11-06 | 2020-02-18 | 深圳微品致远信息科技有限公司 | Method and device for predicting departure taxi time based on deep neural network |
| CN111552178A (en) * | 2020-04-23 | 2020-08-18 | 桂林电子科技大学 | Method for controlling waiting release of aircraft stand with controllable repeat request time interval |
| CN111552178B (en) * | 2020-04-23 | 2022-08-19 | 桂林电子科技大学 | Aircraft stand waiting and pushing control method with controllable repeat request time interval |
| CN112287561A (en) * | 2020-11-18 | 2021-01-29 | 杨媛媛 | Sliding path optimization index solving method based on Monte Carlo algorithm |
| CN113096449A (en) * | 2021-03-25 | 2021-07-09 | 上海行沙科技有限公司 | Flight big data-based shutdown position resource arrangement method and system |
| CN114783212A (en) * | 2022-03-29 | 2022-07-22 | 南京航空航天大学 | Method for constructing model feature set for prediction of departure taxi time of aircraft in busy airport |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846523A (en) | A kind of flight for putting forth coasting time dynamic prediction method based on Bayesian network | |
| CN110472846B (en) | Optimal estimation and uncertainty method for thermal hydraulic safety analysis of nuclear power plant | |
| CN109272138B (en) | Sorting equipment cell number optimization method and device | |
| Steiger et al. | An improved batch means procedure for simulation output analysis | |
| CN107944695A (en) | One kind is based on the probabilistic chaining key setting buffers method of production | |
| Veeger et al. | Generating cycle time-throughput curves using effective process time based aggregate modeling | |
| CN107506444B (en) | Machine learning system associated with interrupted track connection | |
| CN111709454B (en) | Multi-wind-field output clustering evaluation method based on optimal copula model | |
| CN114066262A (en) | Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium | |
| Etman et al. | Aggregate modeling of semiconductor equipment using effective process times | |
| CN111191843A (en) | Airport delay prediction method based on time sequence network propagation dynamic equation | |
| CN114781704A (en) | Flight delay prediction method based on station-passing flight guarantee process | |
| CN106599541B (en) | A kind of structure and parameter on-line identification method of dynamic power load model | |
| JP5501283B2 (en) | Maximum connection number estimation apparatus, method and program | |
| Banerjee et al. | A hybrid statistical method for accurate prediction of supplier delivery times of aircraft engine parts | |
| CN102123053B (en) | Method for analyzing performance of multi-class closed fork-join queuing network based on horizontal decomposition | |
| CN111445079B (en) | Multi-fidelity simulation optimization method and equipment applied to workshop plan production | |
| CN113139344A (en) | Civil aircraft operation risk assessment method oriented to multiple failure modes | |
| Picano et al. | Passengers demand forecasting based on chaos theory | |
| Chang | Risk-controlled product mix planning in semiconductor manufacturing using simulation optimization | |
| CN102158357B (en) | Method for analyzing performances of single closed fork-join queuing network based on horizontal decomposition | |
| Madlenak et al. | Reliability enhancement using optimization analysis | |
| Ghaleb et al. | Using fuzzy logic and discrete event simulation to enhance production lines performance: case study | |
| Siddique et al. | Lead time prediction for sheeter machine production in a paper conversion industry | |
| D'Innocenzo et al. | Verification of temporal properties on hybrid automata by simulation relations |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181120 |
|
| WD01 | Invention patent application deemed withdrawn after publication |