Abstract: In order to improve the local search capacity of genetic algorithms (GAs) in solving the aircraft arrival sequencing and scheduling problem in a multi-runway system, a mixed-integer zero-one quadratic programming model for the problem was built. Then, the proposition was proved that aircrafts of the same type should land on each runway in the order of their expected time of arrival under the first-come first-served policy, and a Memetic algorithm that couples the local improvement algorithm with the genetic algorithm was designed to solve the model. The computational results on several publicly available test problems show that the worst solutions found over ten replications of the Memetic algorithm are not inferior to the best solutions found by some existing GAs, and that the execution time is only 0.17 s for the problem involving 20 aircrafts and 5 runways, and hence is more suitable for real time application compared with the exact algorithm.