Characterization and prediction of air traffic delays

This paper presents a new class of models for predicting air traffic delays. The proposed models consider both temporal and spatial (that is, network) delay states as explanatory variables, and use Random Forest algorithms to predict departure delays 2–24 h in the future. In addition to local delay variables that describe the arrival or departure delay states of the most influential airports and links (origin–destination pairs) in the network, new network delay variables that characterize the global delay state of the entire National Airspace System at the time of prediction are proposed. The paper analyzes the performance of the proposed prediction models in both classifying delays as above or below a certain threshold, as well as predicting delay values. The models are trained and validated on operational data from 2007 and 2008, and are evaluated using the 100 most-delayed links in the system. The results show that for a 2-h forecast horizon, the average test error over these 100 links is 19% when classifying delays as above or below 60 min. Similarly, the average over these 100 links of the median test error is found to be 21 min when predicting departure delays for a 2-h forecast horizon. The effects of changes in the classification threshold and forecast horizon on prediction performance are studied.     

Evaluating the assumption of independent turning probabilities

Several urban traffic models make the convenient assumption that turning probabilities are independent, meaning that the probability of turning right (or left or going straight through) at the downstream intersection is the same for all travelers on that roadway, regardless of their origin or destination. In reality most travelers make turns according to planned routes from origins to destinations. The research reported here identifies and quantifies the deviations that result from this assumption of independent turning probabilities.An analysis of this type requires a set of reasonably realistic “original” route flows, which were obtained by a static user-equilibrium traffic assignment and an entropy maximization condition for most likely route flows. These flows are compared with those route flows resulting from the Assumption of Independent Turning Probabilities (ITP). A small subnetwork of 3 km by 5 km in Tucson, Arizona, was chosen as a case study. An overall “typical ratio” of 2.2 between original route flows and ITP route flows was obtained. Aggregating route flows to origin–destination flows led to an overall “typical ratio” of 1.7. Such deviations are particularly high for routes that go back-and-forth, reaching a ratio of more than 3 in certain time periods. Substantial deviations for origins and destinations that are on the same border of the subnetwork are also observed in the analyses. In addition, under the ITP assumption, morning rush hour traffic peaking is the same in all directions, while in the original flows some directions do not exhibit a peak in the morning rush hour period. Overall, the conclusion of the paper is that the assumption of independent turning probabilities leads to substantial deviations both at the route level and at the origin–destination level, even for such a small network of the case study. These deviations are particularly detrimental when a network is being modeled and studied for route-based measures of effectiveness such as the number and types of routes passing a point – for monitoring specified vehicles and/or managing detouring strategies.     

Automatic identification of risky weather objects in line of flight (AIRWOLF)

Adverse weather conditions are hazardous to flight and contribute to re-routes and delays. This has a negative impact on the National Airspace System (NAS) due to reduced capacity and increased cost. In today’s air traffic control (ATC) system there is no automated weather information for air traffic management decision-support systems. There are also no automatic weather decision-support tools at the air traffic controller workstation. As a result, air traffic operators must integrate weather information and traffic information manually while making decisions. The vision in the Next Generation Air Transportation System (NextGen) includes new automation concepts with an integration of weather information and decision-making tools. Weather-sensitive traffic flow algorithms could automatically handle re-routes around weather affected areas; this would optimize the capacity during adverse conditions. In this paper, we outline a weather probe concept called automatic identification of risky weather objects in line of flight (AIRWOLF). The AIRWOLF operates in two steps: (a) derivation of polygons and weather objects from grid-based weather data and (b) subsequent identification of risky weather objects that conflict with an aircraft’s line of flight. We discuss how the AIRWOLF concept could increase capacity and safety while reducing pilot and air traffic operator workload. This could translate to reduced weather-related delays and reduced operating costs in the future NAS.     

Reverse time-restricted shortest paths: Application to air traffic management

In this paper, we consider a particular class of network flow problems that seeks a shortest path, if it exists, between a source node s and a destination node d in a connected digraph, such that we arrive at node d at a specified time τ while leaving node s no earlier than a lower-bounding time LB, and where the availability of each network link is time-dependent in the sense that it can be traversed only during specified intervals of time. We refer to this problem as the reverse time-restricted shortest path problem (RTSP), and it arises, for example, in the context of generating flight plans within air traffic management approaches under severe convective weather conditions. We show that this problem is NP-hard in general, but is polynomially solvable under a special regularity condition. A pseudo-polynomial time dynamic programming algorithm is developed to solve Problem RTSP, along with an effective heap implementation strategy. Computational results using real flight generation test cases as well as random simulated problems are presented.     

Deconstructing delay: A non-parametric approach to analyzing delay changes in single server queuing systems

This paper introduces an empirically driven, non-parametric method to isolate and estimate the effects that changes in demand and changes in throughput have on delay – in particular, arrival and departure flight delay at airport runways. Classic queuing concepts were used to develop a method by which an intermediate, or counterfactual, queuing scenario could be constructed, to isolate the delay effects due to shifts in demand and throughput. This method includes the development of a stochastic throughput function that is based entirely on data and has three key features. Firstly, the function relies on non-parametric, empirically-based probability distributions of throughput counts. Secondly, facility capacity needs not be explicitly defined, as it is implicitly included in the probability distributions of throughput. Thirdly, the throughput performance function preserves the effect of factors that cause capacity (and, therefore, throughput) to fluctuate over a given period. Temporal sequences of high, moderate, and low capacity are maintained between the observed and counterfactual scenarios. The method was applied to a case study of the three major New York area airports of LaGuardia (LGA), Newark Liberty (EWR), and John F. Kennedy (JFK), using operational data extracted from the Federal Aviation Administration’s (FAA’s) Aviation System Performance Metrics (ASPM) database. The focus was on the peak summer travel seasons of 2006 and 2007, as these airports experienced record levels of delay in 2007. The results indicate that decreases in both demand and throughput were experienced at LGA and EWR, although the decreases in throughput had more significant effects on operational delays as they increased overall at these airports. At JFK, the increase in departure throughput was not sufficient to offset the increase in departure demands. For arrivals, demand increased and throughput decreased. These trends caused a significant growth in delay at JFK between 2006 and 2007.     

Accuracy of reinforcement learning algorithms for predicting aircraft taxi-out times: A case-study of Tampa Bay departures

Taxi-out delay is a significant portion of the block time of a flight. Uncertainty in taxi-out times reduces predictability of arrival times at the destination. This in turn results in inefficient use of airline resources such as aircraft, crew, and ground personnel. Taxi-out time prediction is also a first step in enabling schedule modifications that would help mitigate congestion and reduce emissions. The dynamically changing operation at the airport makes it difficult to accurately predict taxi-out time. In this paper we investigate the accuracy of taxi out time prediction using a nonparametric reinforcement learning (RL) based method, set in the probabilistic framework of stochastic dynamic programming. A case-study of Tampa International Airport (TPA) shows that on an average, with 93.7% probability, on any given day, our predicted mean taxi-out time for any given quarter, matches the actual mean taxi-out time for the same quarter with a standard error of 1.5 min. Also, for individual flights, the taxi-out time of 81% of them were predicted accurately within a standard error of 2 min. The predictions were done 15 min before gate departure. Gate OUT, wheels OFF, wheels ON, and gate IN (OOOI) data available in the Aviation System Performance Metric (ASPM) database maintained by the Federal Aviation Administration (FAA) was used to model and analyze the problem. The prediction accuracy is high even without the use of detailed track data.     

Weather impact on containership routing in closed seas: A chance-constraint optimization approach

Weather conditions have a strong effect on the operation of vessels and unavoidably influence total time at sea and associated transportation costs. The velocity and direction of the wind in particular may considerably affect travel speed of vessels and therefore the reliability of scheduled maritime services. This paper considers weather effects in containership routing; a stochastic model is developed for determining optimal routes for a homogeneous fleet performing pick-ups and deliveries of containers between a hub and several spoke ports, while incorporating travel time uncertainties attributed to the weather. The problem is originally formulated as a chance-constrained variant of the vehicle routing problem with simultaneous pick-ups and deliveries and time constraints and solved using a genetic algorithm. The model is implemented to a network of island ports of the Aegean Sea. Results on the application of algorithm reveal that a small fleet is sufficient enough to serve network’s islands, under the influence of minor delays. A sensitivity analysis based on alternative scenarios in the problem’s parameters, leads to encouraging conclusions with respect to the efficiency and robustness of the algorithm.     

Route Choice Behaviour of Freeway Travellers Under Real-time Traffic Information Provision – Application of the Best Route and the Habitual Route Choice Mechanisms

Abstract

This paper investigates route choice behaviour on freeways between Taipei and Taichung in Taiwan under the provision of real-time traffic information. Two types of route choice selection rules (the best-route and habitual-route) are analysed using ordered probit models to identify the major influences on freeway travellers’ route choice behaviour. The level of service associated with each route is defined as a generalised cost saving (GCS) and specified non-linearly with a threshold inherent to travellers. The marginal (dis)utility thresholds in the ‘best’ and ‘habitual’ behaviour models are identified through a goodness-of-fit grid. The results confirm that the thresholds for changing the inertia behaviour of drivers should be larger than the ones for choosing the best routes. In addition, the drivers are more likely to choose either the best or the habitual routes once the GCS are greater than the identified threshold values.     

航班快速恢复建模与仿真研究

随着民用航空的发展与竞争,航班延误不仅影响航空飞行的安全与正常,更与航空公司的运营效率、运营成本及乘客利益息息相关。针对某一恶劣天气影响,对某公司受影响航班进行重新调配,考虑到航班的备降、盘旋等待、延误、取消等多种状态,以总成本最小为目标函数,建立航班快速恢复模型,通过MATLAB运用遗传算法设计航班恢复算法进行求解,得出最经济的航班恢复方案。     

Adaptive prediction of flight time uncertainty for ground-based 4D trajectory management

An adaptive prediction model of level flight time uncertainty is derived as a function of flight and meteorological conditions, and its effectiveness for ground-based 4D trajectory management is discussed. Flight time uncertainty inevitably increases because of fluctuations in meteorological conditions, even though the Mach number, flight altitude and direction are controlled constant. Actual flight data collected using the secondary surveillance radar Mode S and numerical weather forecasts are processed to obtain a large collection of flight time error and flight and meteorological conditions. Through the law of uncertainty propagation, an adaptive prediction model of flight time uncertainty is derived as a function of the Mach number, flight distance, wind, and temperature. The coefficients of the adaptive prediction model is determined through cluster analysis and linear regression analysis. It is clearly demonstrated that the proposed adaptive prediction model can estimate the flight time uncertainty without underestimation or overestimation, even under moderate or severe weather conditions. The proposed adaptive prediction is able to improve both safety and efficiency of 4D trajectory management simultaneously.     

Monitoring Aircraft Turnaround Operations – Framework Development,Application and Implications for Airline Operations

Abstract

A real-time operation monitoring system – Aircraft Turnaround Monitoring System – is developed based on a system framework to monitor aircraft turnaround operations at an airport. Mobile computing devices (PDAs) and wireless network technology General Packet Radio Service (GPRS) are used to implement the real-time monitoring system for an airline. System implementation and test results indicate that real-time operation monitoring can potentially reduce delays occurring from airline operations. Proactive measures can be taken immediately by ground handling staff to reduce delays, once the risk of delays and potential delay propagation is identified. The availability of detailed operating data can help airlines identify the root delay causes from complex connections among aircraft, flight/cabin crew and passengers. In addition, these operating data also shed some light on the future development of aircraft routing algorithms in order to consider explicitly stochastic disruptions and delay propagation in airline schedule planning.     

A math-heuristic algorithm for the integrated air service recovery

A sophisticated flight schedule might be easily disrupted due to adverse weather, aircraft mechanical failures, crew absences, etc. Airlines incur huge costs stemming from such flight schedule disruptions in addition to the serious inconveniences experienced by passengers. Therefore, an efficient recovery solution that simultaneously decreases an airline's recovery cost while simultaneously mitigating passenger dissatisfaction is of great importance to the airline industry. In this paper, we study the integrated airline service recovery problem in which the aircraft and passenger schedule recovery problems are simultaneously addressed, with the objective of minimizing aircraft recovery and operating costs, passenger itinerary delay cost, and passenger itinerary cancellation cost.Recognizing the inherent difficulty in modeling the integrated airline service recovery problem within a single formulation (due to its huge solution space and quick response requirement), we propose a three-stage sequential math-heuristic framework to efficiently solve this problem, wherein the flight schedules and aircraft rotations are recovered in the first stage, Then, a flight rescheduling problem and passenger schedule recovery problems are iteratively solved in the next two stages. Time-space network flow representations, along with mixed-integer programming formulations, and algorithms that take advantages of the underlying problem structures, are proposed for each of three stages. This algorithm was tested on realistic data provided by the ROADEF 2009 challenge and the computational results reveal that our algorithm generated the best solution in nearly 72% of the test instances, and a near-optimal solution was achieved in the remaining instances within an acceptable timeframe. Furthermore, we also ran additional computational runs to explore the underlying characteristics of the proposed algorithm, and the recorded insights can serve as a useful guide during practical implementations of this algorithm.     

Airport taxi planning: Lagrangian decomposition

The airport taxi planning (TP) module is a decision tool intended to guide airport surface management operations. TP is defined by a flow network optimization model that represents flight ground movements and improves aircraft taxiing routes and schedules during periods of aircraft congestion. TP is not intended to operate as a stand‐alone tool for airport operations management: on the contrary, it must be used in conjunction with existing departing and arriving traffic tools and overseen by the taxi planner of the airport, also known as the aircraft ground controller. TP must be flexible in order to accommodate changing inputs while maintaining consistent routes and schedules already delivered from past executions. Within this dynamic environment, the execution time of TP may not exceed a few minutes. Classic methods for solving binary multi‐commodity flow networks with side constraints are not efficient enough; therefore, a Lagrangian decomposition methodology has been adapted to solve it. We demonstrate TP Lagrangian decomposition using actual data from the Madrid‐Barajas Airport. Copyright © 2011 John Wiley & Sons, Ltd.     

Predicting the operational acceptance of airborne flight reroute requests using data mining

For tools that generate more efficient flight routes or reroute advisories, it is important to ensure compatibility of automation and autonomy decisions with human objectives so as to ensure acceptability by the human operators. In this paper, the authors developed a proof of concept predictor of operational acceptability for route changes during a flight. Such a capability could have applications in automation tools that identify more efficient routes around airspace impacted by weather or congestion and that better meet airline preferences. The predictor is based on applying data mining techniques, including logistic regression, a decision tree, a support vector machine, a random forest and Adaptive Boost, to historical flight plan amendment data reported during operations and field experiments. Cross validation was used for model development, while nested cross validation was used to validate the models. The model found to have the best performance in predicting air traffic controller acceptance or rejection of a route change, using the available data from Fort Worth Air Traffic Control Center and its adjacent Centers, was the random forest, with an F-score of 0.77. This result indicates that the operational acceptance of reroute requests does indeed have some level of predictability, and that, with suitable data, models can be trained to predict the operational acceptability of reroute requests. Such models may ultimately be used to inform route selection by decision support tools, contributing to the development of increasingly autonomous systems that are capable of routing aircraft with less human input than is currently the case.     

Transit passenger response: Short and long term elasticities using time series analysis

This paper investigates the effects of price and service changes on transit ridership. The concept of elasticity is introduced and the traditional methods for estimating elasticities are discussed. In this paper an extra dimension is added by investigating short and long term elasticities. Time series analysis, developed by Box and Jenkins is chosen for the analysis. The Box and Jenkins methodology is applied to a monthly time series of average weekday ridership on the Chicago Transit Authority (CTA) rail system. Four categories of explanatory variables are investigated: fare on the CTA rail system, service provided on the CTA rail system, cost of car trips and weather effects. The effects of gas prices and rail service were found to be significant; however the results indicate a twelve month delay before service changes influence ridership. The effect of transit fares was found to be insignificant, indicating that both the short and long term fare elasticities are zero.     

A finite mixture model of vehicle-to-vehicle and day-to-day variability of traffic network travel times

This study proposes an approach to modeling the effects of daily roadway conditions on travel time variability using a finite mixture model based on the Gamma–Gamma (GG) distribution. The GG distribution is a compound distribution derived from the product of two Gamma random variates, which represent vehicle-to-vehicle and day-to-day variability, respectively. It provides a systematic way of investigating different variability dimensions reflected in travel time data. To identify the underlying distribution of each type of variability, this study first decomposes a mixture of Gamma–Gamma models into two separate Gamma mixture modeling problems and estimates the respective parameters using the Expectation–Maximization (EM) algorithm. The proposed methodology is demonstrated using simulated vehicle trajectories produced under daily scenarios constructed from historical weather and accident data. The parameter estimation results suggest that day-to-day variability exhibits clear heterogeneity under different weather conditions: clear versus rainy or snowy days, whereas the same weather conditions have little impact on vehicle-to-vehicle variability. Next, a two-component Gamma–Gamma mixture model is specified. The results of the distribution fitting show that the mixture model provides better fits to travel delay observations than the standard (one-component) Gamma–Gamma model. The proposed method, the application of the compound Gamma distribution combined with a mixture modeling approach, provides a powerful and flexible tool to capture not only different types of variability—vehicle-to-vehicle and day-to-day variability—but also the unobserved heterogeneity within these variability types, thereby allowing the modeling of the underlying distributions of individual travel delays across different days with varying roadway disruption levels in a more effective and systematic way.     

Estimating freeway route travel time distributions with consideration to time‐of‐day,inclement weather,and traffic incidents

This paper develops an efficient probabilistic model for estimating route travel time variability, incorporating factors of time‐of‐day, inclement weather, and traffic incidents. Estimating the route travel time distribution from historical link travel time data is challenging owing to the interactions among upstream and downstream links. Upon creating conditional probability function for each link travel time, we applied Monte Carlo simulation to estimate the total travel time from origin to destination. A numerical example of three alternative routes in the City of Buffalo shows several implications. The study found that weather conditions, except for snow, incur minor impact on off‐peak and weekend travel time, whereas peak travel times suffer great variations under different weather conditions. On top of that, inclement weather exacerbates route travel time reliability, even when mean travel time increases moderately. The computation time of the proposed model is linearly correlated to the number of links in a route. Therefore, this model can be used to obtain all the origin to destination travel time distributions in an urban region. Further, this study also validates the well‐known near‐linear relation between the standard deviation of travel time per unit distance and the corresponding mean value under different weather conditions. Copyright © 2016 John Wiley & Sons, Ltd.     

Development of network restructuring models for improved air traffic forecasts

Current air traffic forecast methods employed by the United States Federal Aviation Administration function under the assumption that the structure of the network of routes operated by airlines will not change; that is, no new routes will be added nor existing ones removed. However, in reality the competitive nature of the airline industry is such that new routes are routinely added between cities possessing significant passenger demand; city-pairs are also removed. Such phenomena generates a gap between the forecasted and actual state of the US Air Transportation System in the long term, providing insufficient situational awareness to major stakeholders and decision-makers in their consideration of major policy and technology changes. To address this gap, we have developed and compared three algorithms that forecast the likelihood of un-connected city-pairs being connected by service in the future, primarily based on the nodal characteristics of airports in the US network. Validation is performed by feeding historical data to each algorithm and then comparing the accuracy and precision of new city-pairs forecasted using knowledge of actual new city-pairs that developed. While an Artificial Neural Network produces superior precision, fitness function and logistic regression algorithms provide good representation of the distribution of new route types as well as greater flexibility for modeling future scenarios. However, these latter two algorithms face difficulty in resolving differences among the large number of ‘spoke’ airports in the network – additional parameters that may be able to differentiate them are currently under review. These insights gained are valuable stepping stones for exploiting knowledge of restructuring in the service route network to improve overall forecasts that drive policy and technology decision-making.     

Defining and measuring service availability for complex transportation networks

Service Availability of a transportation system is a measure of a performance that has been generally defined according to the reliability and maintainability terms of mean-time-before-failure and mean-time-to-restore, as borrowed from the aerospace/defense industry. While such definitions correctly describe the availability of a system and its equipment to function they do not directly measure the percent of designed and scheduled service available for passenger use. For the more complex transportation systems having multiple tracks and routes, fleets of vehicles, more than two stations and more than one mode of service there are needs for definitions that account for isolated failures that partially interrupt or delay service. Successful definitions of service availability have been based on data that is easily and directly entered in the operating log or automatically collected by Automatic Train Supervision (ATS) system and reports generated by software. The following paper first defines measures of service availability in current use and analyzes exact and approximation methods for data collection and computation. Second, the paper postulates and explores classical and new definitions of service availability applicable for complex networks such as Personal Rapid Transit (PRT). Insight is provided for choosing a suitable definition based on the type of transportation network.