Development and use of Air Passenger Trip Tables in SCAG's 2016 Regional Transportation Plan
Corresponding Author: Patrick Coleman, AECOM
Presented By: Patrick Coleman, AECOM
Commercial air passenger trips by traffic analysis zone (TAZ) for seven commercial airports in Southern California were developed using regression analyses to establish relationships between zonal characteristics and observed air travel trip-making, based on air passenger survey data. Airport trip tables were developed for four trip purposes typically used in analyzing ground access air passenger trips: resident business, resident non-business, visitor business, and visitor non-business.
Trip generation rates for 16 subregions within Southern California were developed for each trip purpose using socio-economic variables included in the region’s Regional Transportation Plan data. Air passenger survey data was compiled and grouped into the four air passenger trip purposes, and regression equations were developed for each of the trip purposes using such variables as number of households, household size, average household income, number of workers, number of high income workers, income weighted population, professional employment, and hospitality employment.
Analysis indicated that, while hospitality employment was in general a good predictor of leisure trips, the regression equation overpredicted air passenger trips to “weekend destinations” within the region, to which many leisure trips are made by automobile. Therefore, separate regressions were performed for the subregions with popular weekend destinations and the rest of the region. In addition, it was observed that zones in Los Angeles County generate more air passenger trips than zones with comparable demographic characteristics in the other counties in the region.
The generated air passenger trips were then distributed from the 16 subregions to the seven airports using a gravity model, matching the total trips by subregion and airport. Gravity models were developed for each of the airports, and friction factors were derived by fitting the trip length frequency distributions to the observed trip lengths from air passenger surveys.
Finally, the mode of arrival for each airport was estimated using an Air Passenger Mode of Arrival Model employing a nested mode choice model for seven modes of access: drop-off, drive, rental, limo, taxi, on-call and public transit. The mode choice model was calibrated separately for each of the seven airports in the region.