|
Air Traffic Management Researchers
& Decision-Making
Testing a Critical Next Generation System Issue
 |
 |
Example of a tool used to analyzing airport capacity. Proper testing is critical to NGATS planning. Credit: Luciad
|
Making critical decisions on transforming the air traffic management system is not an easy task, and must be based on solid research and testing to ensure that changes truly result in improvements to the system. But what does "solid" mean? When the Next Generation System is in place, how will decisions have been made on vital issues, such as aircraft separation, automated routing, or jet noise? Through results achieved through simulated testing? Or those that passed a "real world" try-out?
The answer: When the Next Generation System becomes operational, all of the changes in the system will have been validated through high-fidelity modeling and simulation and eventually through flight operational demonstrations to confirm that sufficient benefits will accrue to justify the investment, according to JPDO planners.
The weighty issue of testing was raised in the latest Air Traffic Control Quarterly, which published the top papers from the US/Europe Seminars on Air Traffic Management Research and Development held last year. Commented Air Traffic Control Quarterly Guest Editor Jeffery A. Schroeder, "This is a matter requiring urgent consideration and careful thought if the objectives of a Next Generation System are to be met in a timely manner."
The Air Traffic Control Association has generously shared the following abstracts from these papers.
Safety Analysis for Advanced Separation Concepts, by J.W. Andrews, H. Erzberger, and J.D. Welch.
A preliminary, fault-tree analysis was performed on a proposed method for using automation, instead of human controllers, to provide separation assurance in certain en route airspace. The combined risk of the four fault types examined was estimated to be 1.8E-12 per hour, which was considerably below the target design of 1.0E-9 per hour.
Human Factors Implications of Continuous Descent Approach Procedures
for Noise Abatement, by H.J. Davison Reynolds, T.G. Reynolds, and R.J. Hansman.
In this paper, it was found that predicting separation for a mix of decelerating and constant-speed aircraft might increase controller workload, which could result in offsetting throughput reductions. An analysis and experiment led to a suggestion of standardizing the deceleration profiles so that the associated intent could ease controller workload.
Simulation of Terminal-Area Flight Management System Arrivals
with Airborne Spacing, by T.J. Callantine, P.U. Lee, J. S. Mercer, E.A. Palmer, and T. Prevot.
A simulation compared performance and workload differences among the conditions in which the cockpit or the ground were aided with tools to maintain spacing for arrivals on flight management
system routes. Spacing accuracy improved when spacing tasks were delegated to the equipped cockpits. Controllers rated their workload higher when they delegated spacing tasks to the aircraft versus when they did not; however, it was suggested that additional tool maturity might allow for increased controller acceptance of airborne spacing.
Quantifying Convective Delay Reduction Benefits for Air Traffic Management Systems, by J.E. Evans, M. Robinson, and S. Allan.
This paper presents the difficulties that have arisen to date on measuring how useful new operational weather systems have been in reducing flight delays. Based on the experience gained, recommendations to evaluate future weather system benefits include interviewing operational users followed by detailed case analyses, analyzing flight tracks before and after system installation, and normalizing delay statistics with a convective weather metric.
The Use of Panel Data Analysis Techniques in Airspace Capacity Estimation, by A. Majumdar, W.Y. Ochieng, G. McAuley, J.M. Lenzi, and C. Lepadatu.
A workload model was derived from simulation data in two dissimilar airspace regions using a cross-sectional time-series analysis. Key components in the model were developed from interviews with controllers. It was found that variables that best predict controller workload are different during peak traffic hours than during non-peak traffic hours.
|
