Taxi Time Prediction at Charlotte Airport Using Fast-Time Simulation and Machine Learning Techniques

Lee, Hanbong

doi:10.2514/6.2015-2272

Cited by 26 publications

(23 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers have previously used different techniques to predict taxi-time: Queuing models, stochastic node-link models, discrete event simulations, and models based on statistical regression [9][10][11][12]. From the perspective of controlling pushbacks at the gate, the focus has been on developing models for the departure process.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the Impact of Uncertainty on Airport Surface Operations

Badrinath

Balakrishnan

Clemons

et al. 2018

2018 Aviation Technology, Integration, and Operations Conference

View full text Add to dashboard Cite

Flights spend significantly more time taxiing on the airport surface during periods when the departure demand exceeds airport capacity, resulting in excessive fuel burn. Departure metering by holding aircraft at the gate during periods of congestion has been shown to yield benefits by lowering the taxi-out time. However, an important aspect of this problem that has not been understood well is the impact of uncertainty in departure demand. Recently, some airlines are beginning to publish an expected time that the flights are ready to pushback, which is referred to as Earliest Off-Block Time (EOBT). Tactical decisions for departure metering need to be made with the EOBT information. However, the EOBT published by airlines is often found to deviate from the actual gate out time without departure metering, which represents an error in the EOBT estimate. Hence, it is important to consider errors in EOBT information while analyzing benefits from departure metering.In this paper, we present a queuing network model to predict aircraft taxi-times on the airport surface. The predictions from the queue model are used for departure metering with NASA's ATD-2 logic that is being used in field trials at Charlotte airport. The framework allows us to quantify the reduction in departure metering benefits due to errors in EOBT information. The analysis reveals that the benefits reduce significantly due to EOBT uncertainty which has important implications for future departure metering applications, such as through the Terminal Flight Data Manager (TFDM) platform.

show abstract

Section: Introductionmentioning

confidence: 99%

Evaluating the Impact of Uncertainty on Airport Surface Operations

Badrinath

Balakrishnan

Clemons

et al. 2018

2018 Aviation Technology, Integration, and Operations Conference

View full text Add to dashboard Cite

show abstract

“…Also, they used the same explanatory variables for different approaches, which included multiple linear regression, least median squared linear regression, Support Vector Regression, M5 model trees, Mamdani fuzzy rule-based systems, and TSK fuzzy rule-based systems, to predict taxi-out times and then compared these approaches [15]. Lee et al used both fast-time simulation and machine-learning techniques to predict taxi-out time and found the prediction method of Support Vector Regression to be better than the linear regression method and the Dead Reckoning method [16].…”

Section: Literature Reviewmentioning

confidence: 99%

Predicting Taxi-Out Time at Congested Airports with Optimization-Based Support Vector Regression Methods

Lian

Zhang

Desai

et al. 2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Accurate prediction of taxi-out time is significant precondition for improving the operationality of the departure process at an airport, as well as reducing the long taxi-out time, congestion, and excessive emission of greenhouse gases. Unfortunately, several of the traditional methods of predicting taxi-out time perform unsatisfactorily at congested airports. This paper describes and tests three of those conventional methods which include Generalized Linear Model, Softmax Regression Model, and Artificial Neural Network method and two improved Support Vector Regression (SVR) approaches based on swarm intelligence algorithm optimization, which include Particle Swarm Optimization (PSO) and Firefly Algorithm. In order to improve the global searching ability of Firefly Algorithm, adaptive step factor and Lévy flight are implemented simultaneously when updating the location function. Six factors are analysed, of which delay is identified as one significant factor in congested airports. Through a series of specific dynamic analyses, a case study of Beijing International Airport (PEK) is tested with historical data. The performance measures show that the proposed two SVR approaches, especially the Improved Firefly Algorithm (IFA) optimization-based SVR method, not only perform as the best modelling measures and accuracy rate compared with the representative forecast models, but also can achieve a better predictive performance when dealing with abnormal taxi-out time states.

show abstract

“…Zhang et al (2010) propose an ordered response model by combining a series of variables and an iterative algorithm for predicting taxi-out time. Considering the uncertainty in taxiing time, fuzzy rule (Chen et al, 2011) and machine learning method (Lee et al, 2015) have been adopted to enhance the robustness of prediction during daily operations. Zhang and Wang (2017) develop a new method relying on and econometrics regression models to estimate taxiing performance.…”

Section: Departure Traffic Managementmentioning

confidence: 99%

Fundamental diagrams of airport surface traffic: Models and applications

Yang

Yin

Han

et al. 2017

Transportation Research Part B: Methodological

View full text Add to dashboard Cite

This paper reveals and explores the flow characteristics of airport surface network on both mesoscopic and macroscopic levels. We propose an efficient modeling approach based on the cell transmission model for simulating the spatio-temporal evolution of flow and congestion on taxiway and apron networks. The existence of link-based fundamental diagram that expresses the functional relationship between link density and flow is demonstrated using empirical data collected in Guangzhou Baiyun airport. The proposed CTM-based network model is shown to be an efficient and accurate method capable of supporting air traffic prediction and decision support. In addition, using both CTM-based simulation and empirical data, we further reveal the existence of an aggregate relationship between traffic density and runway throughput, which is referred to as macroscopic fundamental diagram (MFD) in the literature of road traffic. The MFD on the airport surface is analyzed in depth, and utilized to devise several robust off-block control strategies under uncertainties, which are shown to significantly outperform existing off-block control methods

show abstract

Taxi Time Prediction at Charlotte Airport Using Fast-Time Simulation and Machine Learning Techniques

Cited by 26 publications

References 20 publications

Evaluating the Impact of Uncertainty on Airport Surface Operations

Evaluating the Impact of Uncertainty on Airport Surface Operations

Predicting Taxi-Out Time at Congested Airports with Optimization-Based Support Vector Regression Methods

Fundamental diagrams of airport surface traffic: Models and applications

Contact Info

Product

Resources

About