The ATCo work shift scheduling problem dealing with incidents in a tactical phase

This paper deals with a variation of the air traffic controller (ATCo) work shift scheduling problem focusing on the tactical phase, in which the plan for the day of operations can be modified according to real-time traffic demand or other possible incidents (one or more ATCos become sick and/or there is an increase in unplanned air traffic), which may lead to a new sectorization and a lower number of available ATCos. We propose a new methodology consisting of two phases. The goal of the first phase is to build an initial possibly infeasible solution, taking into account the sectors that have been closed or opened in the new sectorization, together with the ATCos available after the incident. In the second phase, we use the simulated annealing (SA) and variable neighborhood search (VNS) metaheuristics to derive a feasible solution in which the available ATCos are used and all the ATCo labor conditions are met. A weight additive objective function is used in this phase to account for the feasibility of the solution but also for the number of changes in the control center at the time the incident happens and the similarity of the derived solution with templates usually used by the network manager operations center.