Bayesian Network Analysis of Flight Delays

Flight delay is one of the most pressing problems in the National Airspace System (NAS). Because of the major economic and operational impacts of flight delay, it is essential for the Federal Aviation Administration to understand the causes of delay and to find ways to reduce delay. Delay is an inherently stochastic phenomenon. Even if all known causal factors could be accounted for, macro-level national airspace system (NAS) delays could not be predicted with certainty from micro-level aircraft information. This paper presents a stochastic model to analyze the major factors that influence flight delay. The model applies a Bayesian network to represent the interactions among factors that affect delay at each of the major flight phases. The model represents the primary contributing factors to delay at each phase, and combines the models for the individual phases into an overall delay propagation model. Two case studies on delays of departure flights from Chicago O’Hare International Airport (ORD) to Hartsfield-Jackson Atlanta International Airport (ATL) and flights from LaGuardia Airport (LGA) to ATL reveal the common structure of system level environmental and human-caused factors combining to affect components of delay, and how these components contribute to the final arrival delay at the destination airport.