1. ¹Imperial College London, London, UK
2. ²University of Southampton, Southampton, UK

Correspondence: AR Brentnall, Institute for Mathematical Sciences, 53 Prince's Gate, Imperial College London, London SW7 2PG, UK. E-mail: a.brentnall@imperial.ac.uk

Received February 2007; Accepted April 2008; Published online 18 June 2008.

Abstract

Although various airport landing sequencing algorithms have been considered in the literature, little work has been done in comparing their effects on Air Traffic Control, especially against first-come first-served (FCFS) runway sequences, the method most widely used in practice. This paper compares a number of such algorithms using a discrete-event simulation model of an airport with a single landing runway. Statistical methods are used to test for effects of sequencing algorithm, delay-sharing strategy, arrival rate and wake-vortex mix. Little benefit to delay, or stability of sequencing advice, is found from advanced sequencing when small changes are made to inputs calibrated to a specific airspace. Advanced sequencing improves landing rate, compared with FCFS sequencing, only when aircraft arrival rate is greater than maximum runway landing rate, and wake-vortex mix is sufficiently varied. Constrained position shifting constraints limit these improvements and it is shown that deterministic optimal techniques may actually be sub-optimal in a dynamic environment. Our main conclusion is that FCFS is a robust method under many conditions.