The ATC Work Shift Scheduling Problem Based on Multistart Simulated Annealing and Regular Expressions

In this paper we propose a new approach to solving the air traffic controller (ATC) work shift scheduling problem. This approach that minimizes the number of ATCs required to cover a given airspace sectoring, while satisfying a set of ATC labor conditions. First, initial feasible solutions are built using a heuristic, and then multistart simulated annealing is used to reach optimal solutions. In the search process, we use regular expressions to check the feasibility of the visited solutions. This provides high testing speed. Once the optimal ATC number is reached, it is used as the initial solution for a new optimization process aimed at balancing the ATC workloads.