Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
The airline crew pairing (ACP) problem is one of the most challenging problems in airline operations. It aims to decide the optimal connections among pairs of flights assigned to flight crews. However, the number of connections grows exponentially with the increasing number of flights. Conventional approaches usually follow a two-stage method, i.e., divide-and-conquer. The flight sequences (pairings) spanning multiple days are firstly generated for each crew. Then, the best pairing set is chosen based on minimum operational costs. When the number of flights is large, it becomes too difficult to generate all feasible pairings and find the optimum. In order to solve large-scale ACP problems efficiently, we propose a novel iterative optimization framework based on monte carlo tree search (MCTS). Thus, the speed of pairing generation and solution accuracy can be improved. To evaluate the performance of the proposed method, we conduct experiments on different scales of real-world instances provided by airline companies. The empirical results show that the proposed approach is capable of producing high-quality solutions, especially in large-scale instances.