Aircraft Taxi Times at U.S. Domestic Airports

The objective of this study was to investigate aircraft taxi-in and taxi-out times at U.S. domestic airports. NAS-wide models do not typically include detailed representations of airport surface operations, and therefore often rely on simplified models of taxi times. We have taken a novel, statistical approach to characterize taxi behavior, which attempts to separate deterministic effects, such as departure queuing delays, from variations that can be treated as noise. This statistical approach provides a distribution of taxi-out and taxi-in times, which result from unmodeled effects such as crew performance variation, weather, and runway configuration. Within this investigation, we created taxi time parameters for numerous combinations of airports, carriers, and equipment types usable for NAS-wide modeling. Since these parameters generate a distribution of values, they are more robust than current taxi time values. The simplicity of our methods allow for easy periodic updating. Since carriers typically do not make drastic year-to-year changes, we believe our results are acceptable for use in near to mid-term NAS-wide analysis.