Shannon Zelinski

An Airspace Concept Evaluation System Characterization Of National Airspace System Delay

This paper presents an analysis of National Airspace System (NAS) delay. An interacting set of models called the Airspace Concept Evaluation System (ACES) was used to simulate one day of NAS-wide air tra‐c. A total of 36 simulations were run. They included nine difierent airport capacity conditions across the NAS and four levels of NAS-wide demand. The analysis of delay results for these 36 simulations shows that delay increases quadratically with increased demand. The delay versus capacity curves have linear trends. However, nonlinearities within these curves expand with increased demand. These nonlinearities suggest that additional factors such as regional concentrations of airports in low capacity conditions due to regional weather do have a signiflcant in∞uence on the results. This analysis is an important flrst step in exploring and understanding the intricacies of the NAS so that further efiorts can be made to improve it. I. Introduction Over the past several decades, the demand for air travel has increased. The National Airspace System (NAS) has evolved with modest advances in weather/wind prediction, faster and quieter airplanes, and by adding new runways and technologies at airports. Each small step plays a part in the attempt to minimize delay as demand continues to grow. However, current technology is fast approaching the point of diminishing returns. New airspace operational concepts will be necessary to reverse this imbalance. These conceptual changes may afiect the entire NAS in unpredictable ways. The Airspace Concept Evaluation System (ACES) is a NAS-wide fast-time simulation developed at NASA Ames Research Center. ACES simulates a model of the NAS with interacting agents for center control, terminal ∞ow management, airports, individual ∞ights, and other NAS elements. These agents pass messages between one another similar to real world communications. This distributed agent based system is designed to emulate the highly unpredictable nature of the NAS, making it a suitable tool to evaluate envisioned airspace concepts. Before new concepts can be evaluated, the tool must be used to evaluate the current NAS situation. This will lead to the development of processes that attempt to better understand the interdependencies between elements making up the NAS. Through these processes, the cascading efiects that interdependencies produce on the system can be better understood. As a flrst step, this study was performed to establish an initial characterization of NAS-wide delay. The goal was to use ACES to simulate a variety of demand and capacity scenarios in the NAS to quantitatively establish their efiects on system-wide delay. The study included thirty-six simulations, each encompassing a 24-hour demand period. The simulations varied in ∞ight demand and airport operational capacity. Four ∞ight demand cases were studied which included a representation of current-day 2002, an approximate doubling of current-day 2002, and two intermediate ∞ight demand schedules. All cases included domestic commercial passenger ∞ights operating between the top 98 US airports. Capacity was varied with nine cases representing difierent airport operational capacities. Two cases represented optimum and worst-case airport states. All airports operated under visual ∞ight rules (VFR) during the simulation period for the optimum case. For the worst case, the 30 benchmark airports 8 operated

Comparing methods for dynamic airspace configuration

This paper compares airspace design solutions for dynamically reconfiguring airspace in response to nominal daily traffic volume fluctuation. Airspace designs from seven algorithmic methods and a representation of current day operations in Kansas City Center were simulated with two times todays demand traffic. A three-configuration scenario was used to represent current day operations. Algorithms used projected unimpeded flight tracks to design initial 24-hour plans to switch between three configurations at predetermined reconfiguration times. At each reconfiguration time, algorithms used updated projected flight tracks to update the subsequent planned configurations. Compared to the baseline, most airspace design methods reduced delay and increased reconfiguration complexity, with similar traffic pattern complexity results. Design updates enabled several methods to as much as half the delay from their original designs. Freeform design methods reduced delay and increased reconfiguration complexity the most.

Effect of Dynamic Sector Boundary Changes on Air Traffic Controllers

The effect of dynamic sector boundary changes on air traffic controller workload was investigated with data from a human-in-the-loop simulation. Multiple boundary changes were made during simulated operations, and controller rating of workload was recorded. Analysis of these data showed an increase of 16.9% in controller workload due to boundary changes. This increased workload was correlated with the number of aircraft handoffs and change in sector volume. There was also a 12.7% increase in average workload due to the changed sector design after boundary changes. This increase was correlated to traffic flow crossing points getting closer to sector boundaries and an increase in the number of flights with short dwell time in a sector. This study has identified some of the factors that affect controller workload when sector boundaries are changed, but more research is needed to better understand their relationships.

A framework for integrating arrival, departure, and surface operations scheduling

This paper proposes a framework for integrating scheduling between arrival, departure, and surface operations to address the drawbacks of domain segregated scheduling. The framework organizes scheduling tasks by time horizon rather than domain. The four-level framework hierarchy includes the configuration schedule, flight schedule, flight schedule update, and schedule conformance. Current NASA research gaps within this framework are discussed and key areas are proposed where future research should focus to facilitate scheduler integration.

Optimizing integrated terminal airspace operations under uncertainty

In the terminal airspace, integrated departures and arrivals have the potential to increase operations efficiency. Recent research has developed genetic-algorithm-based schedulers for integrated arrival and departure operations under uncertainty. This paper presents an alternate method using a machine jobshop scheduling formulation to model the integrated airspace operations. A multistage stochastic programming approach is chosen to formulate the problem and candidate solutions are obtained by solving sample average approximation problems with finite sample size. Because approximate solutions are computed, the proposed algorithm incorporates the computation of statistical bounds to estimate the optimality of the candidate solutions. A proof-of-concept study is conducted on a baseline implementation of a simple problem considering a fleet mix of 14 aircraft evolving in a model of the Los Angeles terminal airspace. A more thorough statistical analysis is also performed to evaluate the impact of the number of scenarios considered in the sampled problem. To handle extensive sampling computations, a multithreading technique is introduced.

Defining dynamic route structure for airspace configuration

This article describes a method for defining route structure from flight tracks. Individual merge and diverge intersections between pairs of flights are identified, clustered, and grouped into nodes of a route structure network. Links are placed between nodes to represent major traffic flows. A parametric analysis determined the algorithm input parameters producing route structures of current day flight plans that are closest to todays airway structure. These parameters are then used to define and analyse the dynamic route structure over the course of a day for current day flight paths. Route structures are also compared between current day flight paths and more user-preferred paths such as great circle and weather avoidance routing.

GPU-based Parallelization for Schedule Optimization with Uncertainty

This paper presents an application of Graphics Processing Units (GPU) technology for speeding up a schedule optimization problem under uncertainty and provides a fast decision support algorithm to solve an air traffic management problem. In terminal airspace, integrated departure and arrival operations using shared resources have the potential to increase operations efficiency. However, results and benefits from integrated operations might be sensitive to flight time uncertainty. In previous work, a scheduling algorithm was proposed for a model of the Los Angeles terminal airspace. Uncertainty was introduced in the flight times and the uncertainty cost computation was handled by Monte Carlo simulations. The original implementation was carried out on sequential processors, but a 30-minute scenario ran in 6.5 hours, which prohibits applying the algorithm in real-time. This paper presents a GPU-based implementation of the scheduling optimization with uncertainty achieving a 637x speedup in Monte Carlo simulations and a 154x speedup for the entire algorithm compared to a sequential implementation. The runtime of the GPU-based code for the same 30-minute scenario is about 2.5 minutes. This significant speedup allows a large range of experiments to be explored and hundreds of simulations to be run. Two types of experiments are designed and they explore different values of traffic densities and arrival-to-departure ratios. The results demonstrate that there exist trade-off solutions between computed delays and number of controller interventions. The variation of total number of aircraft showed a larger impact on the controller’s workload than the variation of arrival-to-departure ratios. When the traffic density is increased, compromise solutions can be identified to reduce the number of controller interventions and achieve low delays.

Simplified dynamic density based capacity estimation

Methods for estimating constant and variable sector capacity based on an airspace complexity metric, simplified dynamic density, are proposed. Simplified dynamic density is a weighted sum of seven traffic components that contribute to airspace complexity. Constant and variable estimates of maximum sector capacity, based on projected flight tracks, are used to constrain the traffic demand in fast-time simulations. Delays and aircraft counts resulting from these methods are compared with those obtained using the capacities in the current system and based on the “5/3 of average sector flight time” rule. Results show that the simplified dynamic density based capacities produce lower system-wide delays and more throughputs, and indicate more predictable air traffic demands during the peak traffic period.

Dynamic stochastic scheduler for integrated arrivals and departures

In terminal airspace, inefficient operations occur frequently due to constrained airspace and uncertainty. Choke points can easily form in the terminal area and therefore reduce the efficiency of the entire National Airspace System. Based on previous work on scheduling of aircraft arrivals and departures with shared fixes in terminal airspace and uncertainty in departure and arrival times, this work extends the previous stochastic scheduler with dynamic capability such that the scheduler can be sequentially applied to air traffic in terminal airspace with a much larger time frame through sliding windows instead of a static 30-minute traffic scenario. Results show that great delay savings can be achieved by using the dynamic stochastic scheduler relative to current air traffic control procedures. With a 30-minute time window, if an aggressive solution is chosen, on average 5.2 hours can be saved in a day in Los Angeles out of the 30% arrivals and 10% departures that are covered in the experiment. The expected value of annual fuel saving would be more than 10 million dollars. However, the cost of potential controller intervention, which results from the uncertainty of estimated departure and arrival times, will increase by 50% on average. If a moderate solution is chosen instead, with the same expected controller intervention as using current procedure, more than four hours delay saving can still be expected. When uncertainty of departure time increases with look-ahead time, experiments show that optimizations with large windows still find better solutions than the ones with small windows when delay saving is moderate. However, when delay saving is high, time-varied uncertainty plays a more important role than window size, where a small window is preferred for finding good solutions.

A graph-based approach to nominal terminal routing

As the air transportation system moves toward trajectory-based operations, there is a greater need for route modeling, especially in the terminal airspace where traffic flows twist and turn and many do not follow any currently published route. This paper presents a graph-based algorithmic method for defining nominal terminal routing from historical flight tracks. First historical traffic is used to generate a directed graph. Then Djikstras shortest path algorithm is used to identify the shortest yet most commonly used paths through the graph. There were many different nominal routes between different engine types due to the distinct differences in engine performance characteristics. A comparison of graph-based shortest path routes to two other methods (manual and k-means) showed how the method could rapidly produce similar distinct dominant terminal area flows. Finding shortest paths from all links of a sufficient weight, rather than just source nodes, identified more route options depicting standard path control techniques in addition to the dominant route. This method may be used to rapidly prototype route models to assess trajectory-based terminal area concepts.