Local Search for Aircraft-Landing Planning

In collaboration with EUROCONTROL (the European Organisation for the Safety of Air Navigation), the considered Aircraft Landing Planning (ALP) problem aims at minimizing delays (with respect to the published airline schedules) while satisfying the separation constraint (which imposes minimum threshold times between planes, ranging from 90 to 240 s). In this study, the landing sequence of the planes has to be determined first, and subsequently their associated landing times and Holding-Stack Patterns (HSPs) needed to meet such landing times. HSPs consist of making a plane wait for its planned landing time by making circular patterns close to the airport. The uncertainty due to winds is taken into account in the simulation procedure (it has an impact on the arrival times). The proposed solution method is a descent local search with restarts. It is quick enough with respect to implementation in real situations as it can be applied within seconds. Furthermore, the obtained results show that the delays can be reduced by approximately 50% on average when compared to a common practice rule.

Automated summary

This study focuses on the Aircraft Landing Planning (ALP) problem, aiming to minimize delays and satisfy separation constraints. By determining the landing sequence, associated landing times, and Holding-Stack Patterns (HSPs), delays can be reduced by approximately 50% on average.