Cockpit crew scheduling using set partitioning problem

Efficient crew scheduling for cockpit crew is important in airline industry due to operational safety and cost associated with the flight duty time. Because of complexity of regulations imposed to the cockpit crew, it is complicated to generate an efficient schedule. Schedule of cockpit crew can be generated through two steps: selecting the flight patterns, and scheduling of them to the specific time horizon. Heuristic method is developed and applied with massive data in a limited time of computation. A set of flight patterns is selected from all possible flight patterns, which are generated by composing the flight leg based on regulations, by using the set partitioning problem with objective function of oversea stay cost. The selected set of flight patterns found at the first step is allocated to 4 week crew schedule to minimize the variance of total flight time assigned to each crew. The crew schedules obtained are evaluated and compared with the ones currently used in one of major airline company.