Non-stationary delay analysis of runway systems

This paper proposes a new approach for the estimation of aircraft delays at airports given time-varying demand and time-dependent processing times. Based on a characterization of performance models for runway systems, the analysis addresses a runway system where arrivals and departures share a common runway. It is assumed that the requests for landings and take-offs can be modeled as independent Poisson processes and that the processing times are generally distributed with operation-dependent rates. The runway is operated according to the first-come-first-serve rule. A stationary backlog-carryover (SBC) approach is developed for the approximation of time-dependent performance measures for this dynamic queueing system with mixed operations. Numerical examples demonstrate that the SBC approximation is reliable for the analysis of runway queues. Because of its simplicity, the approach is numerically stable and fast.