Modeling NRPs with Soft and Reified Constraints

A nurse rostering problem (NRP) deals with the shifts arrangements of nursing staff in the daily operation of a given health care center. This problem is known to be complex mainly due to its combinatorial nature that commonly requires exponential time to be solved. The literature presents a large list of works devoted to solve NRPs, however most of them are focused on the performance of results of well-known NRP instances, and just a few address real cases. In this paper we present a study of real nurse rostering, which particularly involves a very uncommon shift pattern due to proper country legal regulations. We introduce a new model including soft constraints that can be efficiently solved by employing reified constraints.