Variable neighborhood search for nurse rostering problems

Nurse rostering problems consist of assigning varying tasks, represented as shift types, to hospital personnel with different skills and work regulations. The goal is to satisfy as many soft constraints and personal preferences as possible while constructing a schedule which meets the required personnel coverage of the hospital over a predefined planning period. Real-world situations are often so constrained that finding a good quality solution requires advanced heuristics to keep the calculation time down. The nurse rostering search algorithms discussed in this paper are not aimed at specific hospitals. On the contrary, the intention is that such algorithms should be applicable across the whole sector. Escaping from local optima can be very hard for the metaheuristics because of the broad variety of constraints. In this paper, we present a variable neighborhood search approach. Hidden parts of the solution space become accessible by applying appropriate problem specific neighborhoods. The method allows for a better exploration of the search space, by combining shortsighted neighborhoods, and very greedy ones. Experiments demonstrate how heuristics and neighborhoods can be assembled for finding good quality schedules within a short amount of calculation time.