A Hybrid Multi-objective Genetic-Particle Swarm Optimization Algorithm for Airline Crew Rostering Problem with Fairness and Satisfaction

Abstract

AbstractWith the continuous development of today’s air transport industry, the size of airline crew and flight volume increase continuously. At the same time, crew scheduling becomes increasingly complex and important. The rationality of flight crew scheduling scheme affects flight operation cost and crew satisfaction. Therefore, this paper focuses on the crew scheduling problem aimed at improving crew satisfaction and fairness of work allocation. First, an improved crew scheduling model is established with constraints. Second, a new hybrid multi-objective genetic algorithm is proposed in which a barebones particle swarm optimization (BBPSO) based mutation operator is fused to the framework of non-dominated genetic algorithms. Finally, the experimental results verify the superiority of proposed algorithm based on the actual data of three routes. Moreover, the scheduling scheme could improve market competitiveness and strengthen operation management.KeywordsCrew rosteringBarebones particle swarm optimizationNondominated sorting genetic algorithmMulti-objective swarm intelligence optimization algorithm