Computing Optimal Schedules for Landing Aircraft

In this paper we describe two approaches for solving the problem of scheduling aircraft landing times. In this problem, the air tra c controller must compute a landing time for each aircraft in the horizon, and must attempt to minimise costs whilst observing a number of hard constraints. The costs are associated with aircraft landing times varying from the preferred landing time. The hard constraints are concerned with preventing aircraft from landing within a prescribed minimum time of each other. Given a certain number of aircraft within the air tra c controllers' horizon at a certain point in time, we develop a heuristic for the problem using a genetic algorithm (GA).We also formulate this problem as a mixed integer program (MIP) and develop a branch and bound algorithm for its solution. We compare the quality of the two solutions obtained through computational tests performed on a wide range of problems.