3rd Conference on Artificial Intelligence Applications to the Environmental Science


A Fuzzy Logic System for Automated Short Term Aviation Weather Forecasts

Tom Hicks, Harris Corporation, Melbourne, FL; and T. Crawford and M. Wilson

Aviation safety and operational efficiency in the National Airspace System depend heavily on accurate and timely forecasts for weather conditions in the airport terminal area. Studies by the Federal Aviation Administration (FAA) have identified the need for short-term aviation forecasts for the airport terminal area that are updated frequently and are focused on the 0-8 hour time period. The prediction of ceiling and visibility conditions has always been a particular challenge. Harris Corporation, provider of WARP and OASIS weather processing systems for the FAA, has a fuzzy logic based forecast system for short term ceiling and visibility forecasts that has demonstrated significant skill over a verification period of almost two years. The forecasts are always current and are based on fuzzy systems that utilize the latest observed conditions and multiple numerical weather forecast models as inputs.

The forecast system produces automated ceiling, visibility, wind and weather forecasts for 465 terminal locations in the continental United States. The forecasts are generated upon arrival of each new surface observation (METAR) and are therefore always current. Real-time verification statistics for almost a two-year period indicate that the fuzzy logic based ceiling and visibility forecasts have shown improvement not only over guidance forecasts from numerical models, but over official National Weather Service forecasts as well. This paper describes the Harris short-term aviation forecast system and provides detailed verification statistics that show its performance relative to other guidance and official forecast products. The paper will also briefly discuss how this tool fits into other aviation-related initiatives at Harris.

extended abstract  Extended Abstract (164K)

Session 4, All aspects of artificial intelligence applications to environmental sciences
Tuesday, 11 February 2003, 2:15 PM-5:15 PM

Previous paper  

Browse or search entire meeting

AMS Home Page