On the Problem of Pairing Aircraft for Closely Spaced Parallel Approaches

The problem of scheduling pairs of aircraft for simultaneous landing onto very closely spaced parallel runways is studied. The pair scheduling problem and its generalization group scheduling problem for simultaneous landing onto parallel runways are formulated and shown to be NP-hard, in general. A genetic pairing scheduler algorithm is developed, capable of handling a wide range of constraints, and used in a real-time human-in-the-loop simulation that was carried out to study the operational concept. Experimental data from these simulations and an extensive set of stress tests are presented and analyzed. Results indicate that while the problem is NP-hard in general, practical instances of the algorithm are not necessarily very hard to solve. As such, the proposed algorithm succeeded in finding and suggesting aircraft pairs that met all the problem constraints and thus were accepted by the controllers in over 97% of the cases. High solution quality, scalable runtime, and flexibility of the proposed algorithm in handling different constraints suggest that it is a suitable candidate for use in a real-time application.