A rolling horizon solution approach for the airline crew pairing problem

The crew pairing problem (CPP) is one step of the airline crew scheduling process. The CPP consists of determining a minimum cost set of feasible pairings such that each flight is covered exactly once and side constraints are satisfied. In the industry, this problem has been traditionally solved by a heuristic three-phase (TP) approach that solves sequentially a daily, a weekly, and a monthly problem. The contribution of this paper is to show that the traditional approach is less efficient to solve the crew pairing problem when the flights schedule is not regular. In fact, we show that to obtain better quality solutions in less computational time it is better to skip the first two phases and directly solve the monthly problem using a rolling horizon (RH) approach based on column generation method. All experiments are tested on real data provided by a major airline company.