Article

Journal of Heuristics

, Volume 15, Issue 4, pp 351-379

A grasp-knapsack hybrid for a nurse-scheduling problem

  • Melissa D. GoodmanAffiliated withSchool of Mathematics, Cardiff University
  • , Kathryn A. DowslandAffiliated withSchool of Mathematics, Cardiff UniversityGower Optimal Algorithms Ltd.School of Computer Science and IT, University of Nottingham
  • , Jonathan M. ThompsonAffiliated withSchool of Mathematics, Cardiff University Email author 

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This paper is concerned with the application of a GRASP approach to a nurse-scheduling problem in which the objective is to optimise a set of preferences subject to a set of binding constraints. The balance between feasibility and optimality is a key issue. This is addressed by using a knapsack model to ensure that the solutions produced by the construction heuristic are easy to repair. Several construction heuristics and neighbourhoods are compared empirically. The best combination is further enhanced by a diversification strategy and a dynamic evaluation criterion. Tests show that it outperforms previously published approaches and finds optimal solutions quickly and consistently.

Keywords

Nurse-scheduling Rostering GRASP Hybrid Knapsack