An analysis of activity scheduling behavior of airport travelers

• Uses a nested logit model to predict activity-travel patterns for airport travelers.
• Employs simulation to predict activity schedules of airport travelers.
• Employs HEV search engine to identify the best tree structure for the NL model.

A nested logit model is presented that can be used to predict the activity pattern of travelers inside an airport based on their socio-demographical characteristics (e.g. gender, age), group size, and travel related information (e.g. number of bags, airport size, and total available time). The availability of such a model enhances representation of the behavior dynamics when simulating airport pedestrian traffic. An internet-based revealed preference survey was used to collect data from persons that visit airports including both travelers and non-travelers. The survey focused on the agenda of subjects’ most recent airport trip, the frequency and attitude concerning certain types of activities they performed inside the airport, and the socio-demographic characteristics of each respondent. Three possible nested logit model structures are analyzed and one has been identified as a plausible and statistically acceptable nested structure. Also, the empirical results demonstrate the applicability of a nested logit model for use in identifying travelers’ activity patterns in an airport.