Taste variations in airport choice models

A mixed multinomial logit model for analyzing choice of departure airport in a multiple-airport system (MAS) is presented. The model aims to capture random taste variations across passengers in response to airport level of service through a set of random coefficients. A case study is carried out for the Baltimore, Maryland-Washington, D.C., MAS. The 1998 Air Passenger Survey database is used to estimate the model. The results indicate significant taste variations in response to flight frequency and airline fare even within smaller segments by both trip purpose and residency status. Analyses of the model provide several insightful results, such as distribution of perceived level of service and time value. In addition, the model is used to simulate the impact of interesting scenarios on market share. Substantial policy implications for airport management are also provided.