فهرست نشریات علمی دسترسی آزاد ایران

In airline planning, tactical decisions related to flight schedule design and fleet assignment play a pivotal role in enhancing operational efficiency and maximizing revenue. On the other hand, ticket pricing, directly influencing market share, is inherently affected by the tactical flight timetable, market uncertainties, and passenger choice behavior. To jointly optimize tactical scheduling decisions and ticket pricing policies, and create optimal interaction between them, this paper proposes a modern data-driven decision-making framework that blends Distributionally Robust Optimization (DRO) with Bi-Level Programming (BLP). In this framework, leveraging historical data and machine learning algorithms, a distributional ambiguity set is first constructed to model uncertainty within the DRO framework. The BLP formulation then captures the interaction between flight scheduling (upper level) and ticket pricing (lower level). Additionally, passengers’ choice behavior is incorporated using a Multinomial Logit (MNL) discrete choice model. To address the computational complexity, a column-and-constraint generation (CCG) algorithm is adopted, enabling model decomposition and enhancing computational efficiency. Finally, the proposed model and solution framework are validated through a case study and a series of numerical experiments. Numerical results demonstrate that, compared to classical approaches, the proposed framework significantly improves market share and airline revenue, ensures robustness against uncertainty and passenger behavior variability, and enhances computational tractability.