A hybrid approach for solving real-world nurse rostering problems

摘要: Nurse rostering is the process of creating a plan for nurse working hours over given time horizon. This problem, most variants which are NP-hard, has been studied extensively many years. Still, practical mostly done manually, often by highly qualified health care personnel. underlines need to address challenges realistic, applied rostering, and implementation advanced methods in commercial software. In this paper, we present an industrial case study software currently used several hospitals other institutions Norway Sweden. The presented problem model rich set hard soft constraints, as required Norwegian hospitals. Our solution approach hybrid: An Iterated Local Search framework that uses Constraint Programming initial construction diversification, Variable Neighborhood Descent iterative improvement. search method shows good results terms quality computation on real world instances. We make these test instances available on-line.