subscribe: Daily Newsletter

 

FAA researcher uses SAS Analytics to forecast delays

0 comments

Delayed air travellers looking to blame someone – anyone – might be interested in what Tony Diana says. A researcher at the Federal Aviation Administration (FAA) in the US, Diana uses SAS Analytics to determine the causes and impacts of flight delays on airlines, airports and the National Airspace System.

Diana's research spotlights delay as a pivotal component of airport and airline performance. Because delays are random, it is difficult for airlines to anticipate them and for airports to manage them before congestion reduces the efficiency of the National Airspace System. The air travel industry and FAA collect massive amounts of data on delays, providing rich opportunities for data mining and forecasting.
"Busy airline and airport executives need to make quick sense of volumes of data," says Diana. "Policymakers also need a clear picture of the situation at hand to assess the potential impact of their decisions. Both rely on analysts to identify trends, provide data that support tactical and strategic decisions, and distil complex details into meaningful relationships."
Diana used SAS to benchmark the efficiency of the largest 35 airports. He uses tools commonly applied in other industries, such as medicine, finance or even psychology, to better understand delays and their factors.
"Airline practitioners know how airports function, but they can't always quantify the likely impact of individual factors on airport delay and congestion. Runway configuration, available capacity, tarmac operations and especially weather conditions at each facility also influence airlines' on-time performance," says Diana.
Diana uses SAS Enterprise Miner, among other SAS technologies, for his analyses.
SAS Enterprise Miner streamlines data mining to create predictive and descriptive models based on analysis of vast amounts of data from across the enterprise. Innovative organisations are using SAS data mining software to detect fraud, minimise risk, anticipate resource demands, increase response rates for marketing campaigns and curb customer attrition.
"The power of SAS is to enable analysts to discover trends and to predict likely occurrences through sophisticated analytical tools. For instance, the variance of block times, the time between a plane leaving the gate at the departure airport and entering the gate at the arrival airport, is a good predictor of schedule reliability and delays," continues Diana.
"In addition, the integration of different perspectives under the same roof is instrumental to understanding complex relationships between demand and capacity, helping analysts provide optimal recommendations to policymakers."