This paper deals with one of the most important components of the airlines planning process – a rotation assignment problem. The problem is formulated as a task to determine which aircrafts should operate given flights under the specific sequencing constraints.
We propose a constraint model that fully describes the problem, suggest some enhancements to improve the efficiency of the model, and also describe the dedicated search strategy. Finally, we encapsulate the model into a local search procedure in the style of Large Neighborhood Search.
The variants of the model are experimentally compared with other techniques using real-life data. The comparison shows that the proposed model achieves significantly better quality of schedules in acceptable time.