Hierarchical Bayesian MCMC Estimation of Airport Operations Counts

Abstract

Accurate counts of aircraft operations at unmonitored or partially-monitored general aviation airports are important due to their role in the allocation of funds for airport development and improvement. While the Federal Aviation Administration annually invests approximately $1B in small commercial and general aviation airports in the United States, fewer than 270 of these 2,950 airports have either full- or part-time air traffic personnel available to register operations counts. Aircraft operations at airports with limited personnel may be counted using temporary acoustic, pneumatic, or video devices, and observations from contract staff. The related sample sizes are inherently small, leading to inaccuracies in the extrapolation of long-term totals. The authors have developed an estimation procedure for use with small sample sizes that employs a Bayesian hierarchical model with a Poisson likelihood function. This procedure is used in conjunction with a counting device that registers operations counts based on aircraft transponder signals (also developed by the authors), and is shown to achieve an accuracy of 2.88% at a general aviation airport staffed with an FAA control tower over a one-month period.