2 1 Ja n 20 22 LINEAR LEXICOGRAPHIC OPTIMIZATION AND PREFERENTIAL BIDDING

This paper addresses the construction of the pilots’ schedules under the preferential bidding system in the airline industry. In this system, the pilots bid on the different activities and the airline constructs the schedules that maximize the scores of the pilots lexicographically: the most senior is served first, then the second one, etc. A sequential approach to solve this problem is natural, and actually quite common: the problem is first solved with the bids of the most senior pilot in the objective function; then it is solved with those of the second most senior without decreasing the score of the most senior, and so on. Each step often involves column generation, requires solving an integer program, and usually takes into account the possible schedules of the less senior crew members approximately, which might compel to time-consuming backtrackings. The known existing exact method for this problem follows the sequential approach. The literature commonly admits that the structure of the problems somehow imposes such an approach. In this work, we show that this belief is actually not completely correct by proposing another exact method that first solves the continuous relaxation of the full problem (with its lexicographic objective) via column generation which provides an upper bound on the optimal value. It then calls an IP solver to compute a lower bound by using only the previously generated columns. Finally, it makes use of the upper and lower bounds to identify columns that might improve the value of the lower bound and calls again the IP solver to compute eventually an optimal solution. The problem of generating new columns is solved using a lexicographic resource-constrained longest path algorithm. With this approach, we show that standard schemes of mathematical programming (such as column generation, bounds, dominance, etc.) can be extended directly to optimization problems with lexicographic objectives. Numerical experiments on industrial instances provided by Air France show that the method can solve these instances in relatively short computational time that is compatible with the operational constraints of the airline.