Abstract: In order to reduce the flight delays caused by airport ground service task scheduling， a multi-objective nonlinear integer optimization model was established to minimize total flight delays and flight delay variance. Ground service task scheduling optimization is an NP-hard problem， so， a double-mutation partheno-genetic algorithm was proposed to solve this problem. The algorithm avoids the phenomenon that the genetic algorithm generates illegal individuals when solving similar problems， and the double-mutation strategy has global search capability. The simulation result showed that the double-mutation partheno-genetic algorithm can solve the ground service scheduling optimization problem including assignment of service teams to airlines and airline-service sequence optimization inside a service team， reduce the total flight delays caused by ground service tasks and avoid long single-flight delay.

Key words: airport ground service, minimize total flight delays, minimize delay variance, multi-objective nonlinear integer optimization model, double-mutation partheno-genetic algorithm