Real-time predictive analytics to estimate air traffic flow rates

A Big data Analytics Infrastructure Testbed (BAIT) has been developed with the concept of providing analytics-as-a-service for algorithm developers to use when creating and testing new capabilities for air traffic management decision support. These capabilities typically require considerable process...

Full description

Saved in:
Bibliographic Details
Main Authors: Kelly, Carol, Craig, Keith, Matthews, Michael
Format: Conference Proceeding
Language:English
Subjects:
Algorithm design and analysis
Big data
Real-time systems
Software algorithms
Storms
Weather forecasting
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A Big data Analytics Infrastructure Testbed (BAIT) has been developed with the concept of providing analytics-as-a-service for algorithm developers to use when creating and testing new capabilities for air traffic management decision support. These capabilities typically require considerable processing power and must be tested and compared with a variety of algorithm modifications and parameter adjustments before they are incorporated into an existing real-time infrastructure. The BAIT infrastructure that has been developed and demonstrated has many applications, including: (1) Easily-configurable testing of multiple versions of the same algorithm simultaneously with automatic load balancing (2) Real-time and faster-than-real-time metrics generation providing the ability to alert air traffic flow managers of upcoming issues such as predicted flow rate limitations due to approaching weather (3) Rapid web-based display generation for the demonstration of real-time analytics and metrics This paper provides a description of the BAIT architecture, an initial use case and computational timing comparison, and a description of BAIT's web browser-based demonstration display.
ISSN:2155-4943
2155-4951
DOI:10.1109/ICNSURV.2015.7121244